Abstract
OBJECTIVE To characterize clinical and laboratory findings in cats with naturally occurring sepsis, emphasizing hemostasis-related findings, and evaluate these variables for associations with patient outcomes.
DESIGN Prospective, observational, clinical study.
ANIMALS 31 cats with sepsis and 33 healthy control cats.
PROCEDURES Data collected included history; clinical signs; results of hematologic, serum biochemical, and hemostatic tests; diagnosis; and outcome (survival vs death during hospitalization or ≤ 30 days after hospital discharge). Differences between cats with and without sepsis and associations between variables of interest and death were analyzed statistically.
RESULTS The sepsis group included cats with pyothorax (n = 10), septic peritonitis (7), panleukopenia virus infection (5), bite wounds (5), abscesses and diffuse cellulitis (3), and pyometra (1). Common clinical abnormalities included dehydration (21 cats), lethargy (21), anorexia (18), pale mucous membranes (15), and dullness (15). Numerous clinicopathologic abnormalities were identified in cats with sepsis; novel findings included metarubricytosis, hypertriglyceridemia, and high circulating muscle enzyme activities. Median activated partial thromboplastin time and plasma D-dimer concentrations were significantly higher, and total protein C and antithrombin activities were significantly lower, in the sepsis group than in healthy control cats. Disseminated intravascular coagulopathy was uncommon (4/22 [18%] cats with sepsis). None of the clinicopathologic abnormalities were significantly associated with death on multivariate analysis.
CONCLUSIONS AND CLINICAL RELEVANCE Cats with sepsis had multiple hematologic, biochemical, and hemostatic abnormalities on hospital admission, including several findings suggestive of hemostatic derangement. Additional research including larger numbers of cats is needed to further investigate these findings and explore associations with outcome.
Sepsis is a severe, potentially fatal clinical syndrome that results from a systemic inflammatory response to infection by bacteria, viruses, protozoa, fungi, or helminths.1,2 In cats, sepsis is associated with substantial illness and high mortality rates, ranging from 7 of 23 (30%) to 11 of 14 (79%) in previous reports.3–8 Sepsis has been described in cats with peritonitis, hepatic abscesses, pyothorax, endocarditis, pyelonephritis, and pyometra.3–8 The severity of sepsis, with its variable etiopathogenesis, makes timely diagnosis and prognostication important. Unfortunately, although quite common, confirming an infection in cats with sepsis is often challenging. Additionally, although naturally occurring sepsis is well studied in human patients, studies in cats with sepsis are limited, and there is no consensus regarding the diagnosis of SIRS in cats.3,6 It is generally accepted that a diagnosis of SIRS can be made on the basis of manifestation of ≥ 2 of the following 4 conditions: body temperature outside of the RI; heart rate outside of the RI; tachypnea or hyperventilation, indicated by decreased Paco2; and leukocytosis, leukopenia, or band neutrophil fraction > 5%.1,3,6 The precise character of these abnormalities and their cutoff values are specifically defined in sepsis in humans, whereas in other species, only some cutoffs have been suggested.1,6
Laboratory findings in cats with severe infections have been described previously; however, cases were not necessarily defined as sepsis in those reports.4,5,7,8 To the best of our knowledge, abnormal laboratory findings in cats with naturally occurring sepsis were investigated previously in 2 studies3,6 only, but hemostasis was not comprehensively evaluated.
One well-documented sepsis-associated phenomenon in human patients, laboratory animals, and dogs is inflammation-induced hemostatic derangement.9–11 This condition is caused by disruption of the delicate balance among the prothrombotic, antithrombotic, and fibrinolytic pathways in severe inflammatory conditions such as sepsis.12–14 Studies15,16 of dogs with sepsis have found hemostatic imbalance, manifested by PT and aPTT greater than their respective RIs, D-dimer and fibrinogen degradation product concentrations above their respective RIs, and antithrombin and protein C activities below their respective RIs within 24 hours after initial examination. Antithrombin and protein C activities below the RIs in dogs with sepsis were associated with death, and improvement in these variables over time was associated with a better outcome.16,17 To the best of our knowledge, comprehensive hemostatic findings in cats with naturally occurring sepsis have not been reported. The purpose of the study reported here was to characterize the clinical and laboratory findings in cats with naturally occurring sepsis, with emphasis on the hemostasis-related findings, and to examine potential associations of these variables with patient outcomes. We hypothesized that cats with sepsis would sustain inflammation-induced hemostatic derangement.
Materials and Methods
Selection of cats and collection of data
Cats were enrolled in the prospective, observational clinical study between January 4, 2012, and November 28, 2014. Cats that were evaluated at the Emergency and Critical Care Department of Hebrew University Veterinary Teaching Hospital and fulfilled the preset criteria for sepsis were eligible for study inclusion. Cats treated with medications affecting hemostasis (eg, glucocorticoids, tranexamic acid, heparin, or NSAIDs) or with blood products during the 3 months prior to this examination were excluded. Cats were defined as having sepsis if infection was confirmed by cytologic, histopathologic, serologic, or bacterial culture results and if ≥ 2 of the following criteria were met: rectal temperature ≥ 39.7°C (103.5°F) or ≤ 37.8°C (100.0°F); heart rate ≥ 225 or ≤ 140 beats/min; respiratory rate ≥ 40 breaths/min; and leukocyte count ≥ 19,500 or ≤ 5,000 cells/μL or presence of a left shift.6,18 For study purposes, a left shift was defined as an absolute band neutrophil count > 1,000 cells/μL or > 10% of all neutrophils in patients with neutropenia, which was a more strict definition than those used in previous studies.6,18
Healthy cats owned by staff of the study facility or cats that were brought to the veterinary teaching hospital's spay and neuter program for elective sur- gery were also eligible for study inclusion. These cats were deemed healthy on the basis of history, physical examination findings, and results of a CBC and serum biochemical analysis, and served as a healthy control group for hemostatic variable analysis.
Demographic and signalment data were collected for all cats. Data collected for cats with sepsis included patient history; physical examination, laboratory, diagnostic imaging (thoracic and abdominal ultrasonography), and surgical findings; diagnosis; and outcome (survival vs death [including euthanasia] during hospitalization and up to 30 days after discharge from the hospital). Patient outcomes after hospital discharge were determined by telephone interviews with cat owners. Informed owner consent was obtained for inclusion of all cats in the study. The study was approved by the Institutional Animal Care and Experimentation Committee of the Hebrew University Veterinary Teaching Hospital.
Laboratory tests
Blood samples (up to approx 3 mL total) were collected for a CBC, serum biochemical analysis, hemostatic evaluation, and blood smear evaluation at the time of the initial examination. Samples were collected into tubes containing potassium-EDTA, and a CBCa was performed ≤ 30 minutes after collection. Blood smears were stainedb with modified Wright staining solution and were evaluated morphologically for presence of neutrophil cytoplasmic toxic changes,19 reactive (or activated) monocytes,20,21 and RBC morphological abnormalities, with manual counts of band neutrophils and nucleated RBCs. The smears were examined by an experienced clinician (IA), who was blinded to the serum biochemical and hemostatic test results and outcomes of the cats. In addition, the platelet count was estimated manually, and smears were examined for platelet clumping to rule out spurious thrombocytopenia in automated counts. When platelet clumping was observed, the automated platelet count was omitted from statistical analyses. When metarubricytosis was observed, the leukocyte count was corrected accordingly, and the corrected count was used in the statistical analyses.
Blood samples for serum biochemical analysis were collected in plain tubes with gel separators, allowed to clot for 30 minutes, and centrifuged ≤ 1 hour after collection. The serum was harvested and analyzedc at 37°C, either immediately (during regular working hours) or after storage at 4°C pending analysis. The analysis of stored samples was performed ≤ 12 hours after collection.
Blood samples for hemostatic tests were collected in tubes containing 3.2% trisodium citrate and centrifuged ≤ 30 minutes after collection to obtain plasma. These citrated plasma samples were either analyzed immediately or stored at −80°C pending analysis. Stored samples were analyzed ≤ 12 months after collection. Coagulometric analyzersd were used for hemostatic measurements. Tests included PT,e aPTT,f fibrinogeng concentration (Clauss method),22 and antithrombin and total protein C activities. Antithrombin activity was measured with a chromogenic substrateh after incubation of the plasma with factor-Xa reagenti in the presence of excess heparin, and the results were reported as percentage activity relative to that of a reference sample (human plasma supplied by the manufacturer). For measurement of total protein C activity, citrated plasma was first incubated with copperhead snake (Agkistrodon contortrix contortrix)–derived protein C activator.j The total protein C activity was quantified with a synthetic chromogenic substratek and reported as percentage activity relative to the manufacturer-supplied human reference plasma. The D-dimer concentration was measured quantitatively by means of a chemistry autoanalyzerl with a latex particle-enhanced immunoturbidimetric assaym containing particles of uniform size that were coated with monoclonal antibodies against the D-dimer epitope.23
Definition of DIC
A diagnosis of DIC was made if hemostatic abnormalities were noted concurrently in the following 3 categories: consumption of clotting factors, reflected by prolonged PT or aPTT, hypofibrinogenemia, or thrombocytopenia (alone or in combination); inhibitor consumption, reflected by low total protein C activity, low antithrombin activity, or both; and excessive fibrinolysis, reflected by high plasma D-dimer concentration.24–26
Statistical methods
The distribution pattern of continuous variables was assessed with the Kolmogorov-Smirnoff test. These variables were compared between groups by use of Student t tests or Mann-Whitney U tests for normally and nonnormally distributed data, respectively. Categorical variables were compared between the 2 groups (healthy cats vs cats with sepsis) by means of χ2 or Fisher exact tests. Associations between quantitative variables were evaluated by Pearson (normally distributed values) or Spearman (nonnormally distributed values) correlation analyses. Associations between variables of interest and death for cats with sepsis were examined by univariate regression analysis, and the variables that were associated with the outcome (P ≤ 0.1) were included in a multivariable regression model. For all tests applied, values of P ≤ 0.05 were considered significant. Analyses were performed with a statistical software package.n
Results
The study included 31 cats with sepsis (18 males [13 neutered and 5 sexually intact] and 13 females [10 spayed and 3 sexually intact]) and 33 healthy control cats (13 males [11 neutered and 2 sexually intact] and 20 females [16 spayed and 4 sexually intact]), with median ages of 4.4 years (range, 0.3 to 13.0 years) and 5.1 years (range, 0.7 to 14.5 years), respectively. The sepsis group included 26 domestic shorthair and 5 purebred cats. The control group included 33 domestic shorthair or domestic longhair cats. There were no significant (P = 0.102, P = 0.095, and P = 0.494, respectively) sex, breed, or age differences between groups. The median hospitalization period was 7 days (range, 1 to 11 days).
At the time of evaluation, median heart rate for the sepsis group was 164 beats/min (range, 100 to 250 beats/min), with tachycardia (heart rate ≥ 225 beats/min) noted in 2 of 31 (6%) cats and bradycardia (heart rate ≤ 140 beats/min) in 5 (16%). Median rectal temperature was 38.1°C (100.6°F; range, 33.5° to 40.9°C [92.3° to 105.6°F]), with hypothermia (rectal temperature ≤ 37.8°C [≤ 100.0°F]) noted in 14 of 31 (45%) cats, and hyperthermia (rectal temperature ≥ 39.7°C [≥ 103.5°F]) in 7 (23%). Median respiratory rate was 49 breaths/min (range, 16 to 120 breaths/min), with tachypnea (respiratory rate ≥ 40 breaths/min) noted in 20 of 31 (65%) cats. Leukocytosis was recorded for 12 of 31 (39%) cats; 12 (39%) had leukopenia, and for the remaining 7 (23%), the WBC count was within the RI. A left shift was detected in 19 of 31 (61%) cats of the sepsis group.
Three of 31 (10%) cats of the sepsis group met 4 of the 4 SIRS criteria, 14 (45%) met 3 of 4 criteria, and 14 (45%) met 2 of 4 criteria. In all cats, one of these criteria was a leukogram abnormality. The underlying etiopathogenesis of sepsis included pyothorax (10/31 [32%]), septic peritonitis (7 [23%]), panleukopenia virus infection (5 [16%]), bite wounds (5 [16%]), abscess and diffuse cellulitis (3 [10%]), and pyometra (1 [3%]). Infection was confirmed on the basis of results for ≥ 1 of the following: bacterial culture (9 cats), cytologic examination (22), histopathologic examination (5), and fecal parvovirus antigen serologic testingo (5). Infection was confirmed by > 1 method in 10 cats. Positive bacterial culture results were accompanied by susceptibility testing; organisms included Pasteurella multocida (5 cats), Fusobacterium spp (2), Prevotella spp (1), and mixed bacterial population (1).
The history and physical examination abnormalities for cats in the sepsis group included dehydration (21/25 [84%] cats with information available), lethargy (21/29 [72%]), anorexia (18/30 [60%]), pale mucous membranes (15/27 [56%]), dullness (15/30 [50%]), dyspnea (12/30 [40%]), signs of abdominal pain (11/30 [37%]), lymphadenopathy (7/31 [23%]), and vomiting (6/30 [20%]). Ultrasonographic findings included pleural effusion (10/31 [32%]), abdominal effusion (7 [23%]), and subcutaneous effusion (1 [3%]). No clinical evidence of bleeding or thromboembolism was noted in any cat. In 13 of 27 (48%) cats for which the data were available, the initial clinical signs of illness were observed by the owners ≤ 2 days prior to the examination, whereas in 12 (44%), the signs were reported as present for 3 to 7 days, and in 2 (7%), clinical signs were identified 8 to 14 days prior to the visit.
Hematologic results for the sepsis group at the time of evaluation were summarized (Table 1). Blood smear evaluation results were recorded for 26 of 31 septic cats. Common leukocyte abnormalities included neutrophil cytoplasmic toxic changes (23/26 [88%] cats; noted as marked in 11/23 [48%]), left shift (19/26 [73%]), lymphopenia (24/30 [80%]), eosinopenia (20/27 [74%]), leukocytosis (12/31 [39%]), leukopenia (12/31 [39%]), monocytosis (16/30 [53%]), and reactive monocytes (7/26 [27%]). Metamyelocytes and myelocytes were identified in 8 of 19 (42%) and 7 of 19 (37%) cats in which left shift was present, respectively. The left shift was characterized as regenerative in 8 of 19 (42%) cats and degenerative in 11 (58%). The RBC abnormalities in cats with sepsis included increased RBC distribution width (17/31 [55%]), microcytosis (17 [55%]), metarubricytosis (11 [35%)], and anemia (9 [29%]). Anemia was invariably nonregenerative and was mostly normochromic (8/9 cats) and normocytic (8/9). The automated platelet count was validated by blood smear examination in 26 of 31 cats, and only these 26 were included in the statistical analyses regarding platelet count. Thrombocytopenia was detected in 9 of 26 (35%) cats.
Results of hematologic analysis for 31 cats with naturally occurring sepsis.
| All cats with sepsis | Survivors | Nonsurvivors | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Analyte | No. of cats | Median (range) | No. (%) < RI | No. (%) > RI | No. of cats | Median (range) | No. (%) < RI | No. (%) > RI | No. of cats | Median (range) | No. (%) < RI | No. (%) > RI | RI |
| PCV (%) | 31 | 30.4 (10.9–46.0) | 9 (29) | 1 (3) | 19 | 30.0 (11.0–46.0) | 6 (32) | 1 (5) | 11 | 32.0 (17.0–40.0) | 3 (27) | 0 (0) | 24–45 |
| Total plasma protein (g/dL) | 29 | 7.0 (5.0–8.4) | 5 (17) | 8 (28) | 17 | 7.0 (5.0–8.4) | 3 (18) | 5 (29) | 11 | 7.2 (5.2–8.4) | 2 (18) | 2 (18) | 5.5–7.5 |
| nRBCs (cells/100 WBCs) | 21 | 0.0 (0.0–18.0) | NA | 11 (35) | 19 | 0.0 (0.0–12.0) | NA | 5 (26) | 11 | 0.0 (0.0–18.0) | NA | 4 (37) | < 1 |
| nRBCs (×103/mm3) | 31 | 0.0 (0.0–10.8) | NA | 11 (35) | 19 | 0.0 (0.0–1.0) | NA | 5 (26) | 11 | 0.0 (0.0–10.8) | NA | 4 (37) | < 0.5 |
| RBCs (×106/mm3) | 31 | 7.6 (2.9–12.8) | 9 (29) | 2 (6) | 19 | 6.6 (2.9–12.8) | 5 (26) | 1 (5) | 11 | 8.5 (3.3–11.1) | 4 (36) | 1 (9) | 6.0–10.1 |
| Hemoglobin (g/dL) | 31 | 10.9 (3.7–16.3) | 8 (26) | 3 (10) | 19 | 10.5 (3.7–10.7) | 5 (26) | 3 (26) | 11 | 10.9 (4.9–16.3) | 3 (27) | 0 (0) | 8.1–14.2 |
| Hct (%) | 31 | 31.2 (10.9–50.9) | 8 (26) | 1 (3) | 19 | 31.1 (10.9–50.9) | 5 (26) | 1 (5) | 11 | 31.8 (13.2–42.9) | 3 (27) | 0 (0) | 27.7–46.8 |
| RBC distribution width (%) | 31 | 21.5 (13.4–26,3) | 4 (13) | 17 (55) | 19 | 18.5 (13.9–24.8) | 2 (11) | 9 (47) | 11 | 21.9 (13.4–21.3) | 1 (9) | 8 (73) | 14.4–19.4 |
| Platelets (×103/mm3) | 26 2 | 11.0 (4.0–628.0) | 9 (35) | 1 (4) | 17 | 212.5 (17.0–628.0) | 6 (33) | 1 (6) | 7 | 210 (4.0–506.0) | 2 (29) | 0 (0.0) | 160–625 |
| Corrected WBCs (×103/mm3) | 31 | 10.0 (0.4–78.8) | 12 (39) | 12 (39) | 19 | 14.4 (0.4–56.4) | 4 (21) | 9 (47) | 11 | 4.4 (0.93–78.8) | 7 (64) | 3 (27) | 6.3–19.6 |
| Band neutrophils (%) | 26 | 15.3 (0.0–62.0) | NA | 22 (85) | 15 | 14.0 (1.0–62.0) | NA | 15 (100) | 11 | 17.0 (0.0–57.0) | NA | 7 (64) | 0–3 |
| Neutrophils (%) | 30 | 40.5 (7.3–92.0) | 9 (32) | 7 (25) | 18 | 66.3 (10.0–89.6) | 4 (25) | 4 (25) | 11 | 35.0 (7.3–92.0) | 4 (36) | 3 (27) | 35–75 |
| Lymphocytes (%) | 30 | 9.5 (2.0–81.2) | 24 (80) | 0 (0) | 18 | 9.5 (2.0–81.2) | 14 (78) | 0 (0) | 11 | 9.0 (2.0–28.0) | 10 (91) | 0 (0) | 20–55 |
| Monocytes (%) | 30 | 5.9 (0.7–68.0) | NA | 16 (53) | 18 | 4.0 (0.7–68.0) | NA | 6 (33) | 11 | 24.0 (1.8–58.2) | NA | 10 (91) | 0–4 |
| Eosinophils (%) | 27 | 1.0 (0.0–33.0) | 20 (74) | 2 (7) | 15 | 1.0 (0.0–14.0) | 14 (93) | 1 (7) | 11 | 3.6 (0.5–33.0) | 5 (46) | 1 (9) | 0–4 |
| Basophils (%) | 27 | 0.0 (0.0–1.0) | NA | 1 (4) | 15 | 0.0 (0.0–1.0) | NA | 1 (7) | 11 | 0.0 (0.0–0.91) | NA | 0 (0) | Rare |
| Band neutrophils (×103/mm3) | 26 | 1.1 (0.0–44.9) | NA | 22 (85) | 15 | 2.0 (0.0–18.9) | NA | 15 (100) | 11 | 0.3 (0.0–44.9) | NA | 7 (64) | Rare |
| Neutrophils (×103/mm3) | 30 | 5.4 (0.1–47.0) | 11 (41) | 10 (37) | 18 | 9.4 (0.1–47.0) | 4 (27) | 7 (47) | 11 | 1.1 (0.1–27.6) | 6 (55) | 3 (27) | 3–13.5 |
| Lymphocytes (×103/mm3) | 30 | 1.0 (0.1–6.0) | 22 (73) | 0 (0) | 18 | 1.0 (0.1–6.0) | 11 (61) | 0 (0) | 11 | 0.4 (0.1–4.9) | 10 (91) | 0 (0) | 2–7.7 |
| Monocytes (×103/mm3) | 30 | 0.5 (0.0–4.7) | NA | 11 (37) | 18 | 0.5 (0.0–4.3) | NA | 7 (39) | 11 | 0.5 (0.0–4.7) | NA | 4 (36) | 0–1 |
| Eosinophils (×103/mm3) | 27 | 0.1 (0.0–2.9) | 19 (70) | 1 (4) | 15 | 0.1 (0.0–1.3) | 11 (73) | 0 (0) | 11 | 0.2 (0.0–2.9) | 7 (64) | 1 (9) | 0.3–1.7 |
| Basophils (×103/mm3) | 27 | 0.0 (0.0–0.1) | NA | 0 (0) | 15 | 0.0 (0.0–0.1) | NA | 0 (0) | 11 | 0.0 (0.0–0.0) | NA | 0 (0) | 0–0.1 |
Laboratory RIs are shown for the hospital where the study was conducted. Cats were identified as having sepsis if infection was confirmed by cytologic, histopathologic, serologic, or bacterial culture results and if ≥ 2 of the following criteria were met: rectal temperature ≥ 39.7°C (103.5°F) or ≤ 37.8°C (100.0°F), heart rate ≥ 225 or ≤ 140 beats/min, respiratory rate ≥ 40 breaths/min, and leukocyte count ≥ 19,500 or ≤ 5,000 cells/μL or presence of a left shift. All samples were obtained at the time of initial examination at the study hospital; not all cats had sufficient sample size for all tests. Survivors and nonsurvivors were defined as cats that did and did not live ≥ 30 days after discharge from the hospital, respectively.
NA = Not applicable. nRBCs = Nucleated RBCs.
Common serum biochemical abnormalities in cats with sepsis were summarized (Table 2). Findings included hyponatremia (30/31 [97%] cats), hypochloridemia (24/31 [77%]), hypoproteinemia (22/30 [73%]), hypertriglyceridemia (22/30 [73%]), high CK activity (20/30 [67%]), hypocalcemia (low total calcium; 19/30 [63%]), hyperbilirubinemia (17/31 [55%]), hypoalbuminemia (14/30 [47%]), low urea concentration (14/30 [47%]), high AST activity (13/30 [43%]), and hypocarbia (low total CO2; 11/29 [38%]). The corrected chloride concentration (median, 120.7 mmol/L [range, 109.5 to 126.7 mmol/L]) was significantly (P < 0.01) higher than the measured chloride concentration (median, 113.1 mmol/L [range, 98.8 to 122.3 mmol/L]).
Selected results of serum biochemical analysis for the same 31 cats as in Table 1.
| All cats with sepsis | Survivors | Nonsurvivors | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Analyte | No. of cats | Median (range) | No. (%) < RI | No. (%) > RI | No. of cats | Median (range) | No. (%) < RI | No. (%) > RI | No. of cats | Median (range) | No. (%) < RI | No. (%) > RI | RI |
| ALT (U/L) | 31 | 52 (10–290) | 6 (19) | 4 (13) | 19 | 44.0 (10.0–290.2) | 4 (21) | 3 (16) | 11 | 61.0 (18.7–248.0) | 2 (18) | 1 (9) | 27–101 |
| Albumin (g/dL) | 30 | 2.3 (0.4–3.5) | 14 (47) | 0 (0) | 19 | 2.3 (0.9–3.1) | 9 (48) | 0 (0) | 10 | 2.4 (0.4–3.5) | 5 (50) | 0 (0) | 2.2–4.6 |
| ALP (U/L) | 30 | 9 (0–272) | 16 (53) | 2 (7) | 19 | 9.0 (2–134) | 10 (53) | 1 (5) | 10 | 8 (0.0–272) | 6 (60) | 1 (10) | 14–71 |
| Amylase (U/L) | 30 | 580 (205.2–2,296.4) | 12 (40) | 1 (3) | 19 | 579.9 (205.2–1,226.3) | 8 (42) | 0 (0) | 10 | 610.8 (294.0–2,296.4) | 3 (30) | 1 (10) | 279–1,254 |
| AST (U/L) | 30 | 53 (16–380) | 0 (0) | 13 (43) | 19 | 50 (16–310) | 0 (0) | 8 (42) | 10 | 56 (31–380) | 0 (0) | 4 (40) | 12–58 |
| Total bilirubin (mg/dL) | 31 | 0.3 (0.0–6.5) | NA | 17 (55) | 19 | 0.2 (0.0–2.7) | NA | 9 (47) | 11 | 0.5 (0.0–6.5) | NA | 7 (63) | 0.00–0.20 |
| Total calcium (mg/dL) | 30 | 8.5 (5.2–10.2) | 19 (63) | 0 (0) | 19 | 8.9 (6.0–9.7) | 12 (63) | 0 (0) | 10 | 7.4 (5.2–10.2) | 7 (70) | 0 (0) | 9–11 |
| Chloride (mmol/L) | 31 | 113.1 (98.8–122.3) | 24 (77) | 0 (0) | 19 | 112.8 (104.4–122.3) | 15 (79) | 0 (0) | 11 | 115.3 (98.8–117.9) | 8 (73) | 0 (0) | 108–118 |
| Cholesterol (mg/dL) | 30 | 115.6 (40.1–213.5) | 4 (13) | 0 (0) | 19 | 115.6 (40.1–140) | 3 (16) | 0 (0) | 10 | 112.9 (47.0–213.5) | 1 (10) | 0 (0) | 90–260 |
| Total CO2 (mmol/L) | 29 | 16.7 (11.5–36.7) | 11 (38) | 7 (24) | 18 | 17.7 (11.5–36.7) | 7 (39) | 6 (33) | 10 | 15.6 (13.1–36.0) | 4 (40) | 1 (10) | 15–21 |
| Creatinine (mg/dL) | 30 | 0.87 (0.3–2.1) | 20 (67) | 0 (0) | 19 | 0.8 (0.3–1.9) | 15 (79) | 0 (0) | 11 | 1.1 (0.6–2.1) | 5 (46) | 0 (0) | 1.1–2.2 |
| CK (U/L) | 30 | 736 (71–39,432) | 1 (3) | 20 (67) | 19 | 736 (71–39,432) | 1 (5) | 13 (68) | 10 | 637.5 (168–20,000) | 6 (60) | 0 (0) | 73–260 |
| GGT (U/L) | 30 | 0.0 (0.0–3.4) | 0 (0) | 0 (0) | 19 | 0.0 (0.0–2.1) | 0 (0) | 0 (0) | 10 | 0.1 (0.0–3.4) | 0 (0) | 0 (0) | 0.0–4.0 |
| Glucose (mg/dL) | 30 | 153 (34.0–262.0) | 2 (7) | 20 (67) | 19 | 151 (71.0–262.0) | 0 (0) | 2 (63) | 10 | 155.1 (34.0–232.0) | 2 (20) | 7 (70) | 63–118 |
| Phosphorus (mg/dL) | 30 | 5.4 (3.4–9.7) | 1 (3) | 9 (30) | 19 | 4.8 (3.4–8.2) | 0 (0) | 3 (16) | 10 | 6.4 (4.3–9.7) | 0 (0) | 6 (60) | 3.2–6.3 |
| Potassium (mmol/L) | 31 | 4.1 (3.1–4.8) | 2 (7) | 1 (3) | 19 | 4.2 (3.8–4.1) | 0 (0) | 1 (5) | 11 | 4.1 (3.1–4.8) | 2 (18) | 0 (0) | 3.6–4.9 |
| Total protein (g/dL) | 30 | 5.7 (2.5–8.8) | 22 (73) | 2 (7) | 19 | 6.0 (2.6–8.8) | 14 (74) | 2 (11) | 10 | 5.4 (2.2–7.7) | 7 (70) | 0 (0) | 6.60–8.40 |
| Sodium (mmol/L) | 31 | 145.8 (134.9–152.4) | 30 (97) | 0 (0) | 19 | 146.4 (135.6–152.4) | 18 (95) | 0 (0) | 11 | 145.1 (134.9–149.6) | 11 (100) | 0 (0) | 145–154 |
| Triglycerides (mg/dL) | 30 | 140.8 (25.5–1,306.3) | 0 (0) | 22 (73) | 19 | 122.4 (25.5–373.3) | 0 (0) | 14 (74) | 10 | 167.8 (44.3–1,306.3 | ) 0 (0) | 7 (70) | 8–80 |
| Urea (mg/dL) | 30 | 44.0 (0.6–134.2) | 14 (47) | 6 (20) | 19 | 37.5 (0.6–110.1) | 10 (53) | 1 (5) | 10 | 64.5 (28.0–134.2) | 3 (0) | 5 (50) | 39–71 |
ALP = Alkaline phosphatase. ALT = Alanine transaminase. GGT = γ-Glutamyltransferase.
See Table 1 for remainder of key.
Hemostatic data on initial evaluation were summarized (Table 3). Cats with sepsis had significantly longer median aPTT (P = 0.001), lower median total protein C and antithrombin activities (P < 0.001 for both), and higher median D-dimer concentration (P = 0.001) than did control cats. The proportion of cats with fibrinogen concentration below or above the laboratory RI (150 to 300 mg/dL) in the sepsis group (5/5 for which fibrinogen concentration was measured) was significantly (P = 0.008) higher than this proportion in the control group (7/23 [30%]). There was a significant (P = 0.01), moderate, positive correlation between total protein C activity and antithrombin activity (r = 0.63). The total protein C activity of cats that died during the study period (median, 37.9% [range, 29.0% to 93.9%]) did not differ significantly (P = 0.06) from the value for cats that survived (median 82.9% [range, 45.5% to 123.0%]).
Comparison of hemostatic test results for the same 31 cats as in Table 1 and a group of 33 healthy control cats.
| Cats with sepsis | Controls | ||||
|---|---|---|---|---|---|
| Analyte | No. of cats | Median (range) | No. of cats | Median (range) | P value |
| PT (s) | 24 | 10.2 (5.9–228.2) | 33 | 9.5 (0.9–23.4) | 0.644 |
| aPTT (s) | 23 | 74.4 (14.6–189.5) | 33 | 45.5 (16.8–134.7) | 0.001 |
| Antithrombin activity (%) | 24 | 138 (70–112) | 33 | 162 (110–172) | < 0.001 |
| Total protein C activity (%) | 15 | 75 (29–123) | 22 | 123 (99–164) | < 0.001 |
| Fibrinogen (mg/dL) | 5 | 301 (76–744) | 23 | 220 (107–575) | 0.727 |
| D-dimer (ng/mL) | 24 | 779 (0–6,960) | 33 | 189 (0–1,175) | < 0.001 |
The P values reflect results of comparisons between median results for cats with sepsis and healthy control cats. Values of P < 0.05 were accepted as significant.
Only 22 of 31 cats with sepsis had hemostatic analyte results available for all 3 categories required for assessment of DIC (clotting factor consumption, inhibitor consumption, and increased fibrinolysis). A diagnosis of DIC was made in 4 of these 22 (18%) cats, and it was not associated with death in the study period.
The 30-day survival rate of cats with sepsis was 19 of 30 (63%). Eleven cats (including 4 that were euthanized at the owners' request because of deterioration) died during this period, and 1 was lost to follow-up. Factors positively associated with death on univariate analysis at P ≤ 0.10 included monocytosis (P = 0.019); severe neutrophil cytoplasmic toxic changes (P = 0.089); hypothermia (P = 0.043); high serum concentrations of urea (P = 0.01), phosphorus (P = 0.022), creatinine (P = 0.063), and bilirubin (P = 0.099); low total protein C activity (P = 0.06); and hyponatremia (P = 0.089). Factors negatively associated with death in the univariate analysis included eosinopenia (P = 0.021) and presence of pleural effusion on ultrasound (P = 0.03). In the final multivariate logistic regression model, none of these associations were significant, and the lowest P value was found for high serum urea concentration (P = 0.056). There was no difference in mortality rates among cats with sepsis that was attributed to different underlying diseases (P = 0.34).
Discussion
Results of the present study showed that cats with naturally occurring sepsis due to various causes have multiple hematologic, biochemical, and hemostatic abnormalities, consistent with a severe systemic condition. The hemostatic findings in these cats were in agreement with the results of other investigations, in which low circulating anticoagulant protein activity was found in dogs with experimentally induced sepsis27 and in human and canine patients with naturally occurring severe sepsis.15,16,28–31 In the present study, we examined several hemostatic analytes that, to our knowledge, have not been previously reported for cats with naturally occurring sepsis, and the result highlighted the complexity of the hemostatic imbalance in such cats, revealing elements of both hyper- and hypocoagulability. Cats with sepsis had significantly longer median aPTT than did healthy control cats, but the median PT, although apparently higher in cats with sepsis, did not differ significantly between the groups. The reason for this finding was not clear. The lack of a significant difference in this variable could have been attributable to low statistical power, resulting from limited sample sizes, or the differences between these findings might have resulted from unique differences in consumption of intrinsic and extrinsic clotting factors or the presence of clotting factor inhibitors in cats with sepsis. In dogs with sepsis, low circulating concentrations of clotting factors have been commonly identified.15,27 One study15 found that the platelet counts did not differ significantly between dogs with sepsis and control dogs. In the present study, absolute thrombocytopenia was noted in 9 of 26 (35%) cats. Prolonged aPTT in dogs with sepsis was significantly associated with death,17,32 and the PT was also prolonged and associated with death, albeit with borderline significance. Interestingly, in our study, although aPTT differed significantly between cats with sepsis and healthy control cats and the PT was slightly higher in cats with sepsis without meeting the cutoff for statistical significance, the aPTT, platelet count, and PT were not associated with death during the study period (ie, during hospitalization and ≤ 30 days after hospital discharge).
Antithrombin and total protein C activities were each significantly lower in cats with sepsis than in healthy control cats in this study. In intergroup comparisons, cats with sepsis had significantly greater median aPTT and plasma D-dimer concentration than did control cats, indicating clotting factor consumption and increased fibrinolysis, respectively. In addition, thrombocytopenia was also noted in 9 of 26 cats with sepsis, and this likely resulted from platelet consumption. The combination of the hemostatic changes in the sepsis group was in line with the basic definition of DIC, that is, an acquired hemostatic disorder that includes clotting factor consumption, inhibitor consumption, and increased fibrinolysis, which is secondary to an inciting illness.24–26,33 Nevertheless, only 4 of 22 (18%) cats that had all required data available met the criteria for a diagnosis of DIC, and presence of DIC was not associated with death in the study period. There is still no consensus definition for DIC in cats. It is noteworthy that in the present study, the definition of DIC used was more strict than that used in a previous study33 of cats with sepsis, and it reflected DIC definitions published for people and dogs.24–26,34,35 Despite evidence of hemostatic derangement and a diagnosis of DIC in some cases, clinical bleeding or signs suggesting thrombosis were not observed in cats in the study.
Total protein C and antithrombin activities were not significantly associated with death of cats with sepsis in this study; however, regarding the former, this nonsignificant (P = 0.06) difference very likely resulted from limited cohort size and low mortality rate, which decreased the power of the statistical analyses. Low protein C activity has typically been associated with a poorer outcome in studies15–17,28–31 of people and dogs with naturally occurring sepsis. In sepsis, protein C depletion is thought to result from degradation by neutrophil elastases, consumption, decreased hepatic synthesis, and potentially, acquired vitamin K deficiency.9 Increasing data suggest that the coagulation cascade is responsible not only for fibrin clot formation and the natural activation of anticoagulant and fibrinolytic pathways, but also for generation of serine proteases, which possess potent proinflammatory properties.36 Antithrombin and protein C also have potent anti-inflammatory properties.9,37,38 Therefore, their depletion can aggravate sepsis-induced inflammation.
Nonregenerative anemia was relatively common in cats with sepsis (9/31 [29%]) in the present study. This has been previously identified in cats with sepsis and is mostly likely due to inflammation,3–6,39–41 although its pathogenesis is complex, possibly including low-grade hemolysis, oxidative damage, inflammation-associated disruption of iron metabolism, decreased RBC life span, inadequate erythropoietin production, and insufficient bone marrow response to erythropoietin.6,7,39–41 Interleukin-6, a proinflammatory cytokine, is associated with anemia of inflammation by inducing the expression of hepcidin, which downregulates expression of the iron export channel ferroportin, thereby blocking iron release from storage and restricting its availability for erythropoiesis.41–43 In the present study, metarubricytosis (identified in 11/31 [35%] cats) was even more common than anemia. To our knowledge, this finding has not been reported previously in cats with sepsis. Generally, metarubricytosis in cats is physiologic and secondary to erythroid regeneration.44 It is also associated with bacterial and viral infections and trauma, particularly with pyothorax,44 which was commonly identified in study cats with sepsis (10/31 [32%]). Low-grade hemolysis cannot be ruled out as a cause of the metarubricytosis, however, because the anemia in this study was invariably nonregenerative, the metarubricytosis might have been pathological and possibly attributable to blood-bone marrow barrier disruption, which can be associated with inflammation, infection, necrosis, hypoxia, thrombosis, infarction, and hyperthermia.44,45 Considering that interleukin-6 expression is reportedly increased in cats with sepsis,3,46 its effects on erythropoiesis and the blood-bone marrow barrier might have contributed to metarubricytosis.47
The leukogram findings in the present study should be interpreted cautiously because leukopenia, leukocytosis, or a left shift were inclusion criteria for defining sepsis. Neither neutropenia nor neutrophilia was associated with death in this study. Cats with sepsis commonly had monocytosis, lymphopenia, and eosinopenia (16/30 [53%], 24/30 [80%], and 20/27 [74%], respectively), which was in agreement with other reports6,46 and likely caused by stress, inflammation, and infection.48–50 Neutrophilic left shift, a hallmark of severe infection and typically a negative prognostic factor in cats,3,5,7,48,51,52 was identified in 19 of 26 (73%) cats and was degenerative in 11 (58%), with evidence of peripheral early myeloid precursors. Nevertheless, these variables were not associated with death in these cats, in contrast to previous findings,51 possibly because the mortality rate in cats of our study was low. Toxic cytoplasmic changes in neutrophils were previously reported for 53% of > 500 ill cats with different diseases52 and were more commonly identified in cats with neutropenia and neutrophilia than in those with normal neutrophil counts. In that study,52 the presence of toxic changes was associated with left shift, longer hospitalization time, and concurrent neutrophil cytoplasmic toxic changes, and left shift with neutropenia or neutrophilia was associated with death.52 Surprisingly, we found no significant association between the presence of these factors and death of cats with sepsis on multivariate analysis, possibly owing to effects of the previously mentioned small sample size and low mortality rate on statistical analyses. The authors consider that, despite these findings, detection of concurrent neutrophil cytoplasmic toxicity and left shift is a negative prognostic factor for such patients, and this should be investigated further.
Hyponatremia, found in 30 of 31 (97%) cats with sepsis in the present study, was mostly accompanied by hypochloridemia, suggesting parallel changes in both ions. However, the corrected chloride concentration was significantly higher than the measured one. Because hyponatremia and hypochloridemia were very common in the sepsis group, a selective increase in serum chloride concentration was highly unlikely. Therefore, this difference between the corrected and measured chloride concentrations resulted from selective decrease in serum sodium concentration, likely due to dilutional acidosis.53 Both hyponatremia and hypochloridemia probably resulted from gastrointestinal (eg, panleukopenia virus infection), renal (eg, pyometra), or third-space (eg, pyothorax and peritonitis) losses, as previously described in cats with sepsis.3 In a previous study3 of SIRS in cats, with and without sepsis, plasma chloride was the only measured analyte significantly associated with death in cats with sepsis and in all cats with SIRS, suggesting that serum chloride concentration is a prognostic marker. Conversely, neither hyponatremia nor hypochloridemia were associated with death in the present study. This discrepancy between studies might have resulted from differences in the etiologies of sepsis and lack of statistical power in the present study.
Low circulating concentrations of total and free calcium have been described in cats with septic peritonitis.3,5,54 Decreased total serum calcium concentration was identified in 19 of 30 (63%) cats of the sepsis group in our study and was potentially associated with hypoalbuminemia,55 noted in 14 of 30 (47%) cats. However, the pathogenesis of sepsis-associated hypocalcemia is not completely understood, and several proposed mechanisms are unproven.54 Twenty-two of 30 (73%) affected cats in our study had hypoproteinemia, and both hypoalbuminemia and hypoproteinemia are common in cats with sepsis.3–7 Hypoalbuminemia was attributed to increased loss, malnutrition, increased capillary permeability and leakage, hepatic dysfunction, decreased production caused by the acute phase response, loss into effusions, and antidiuretic hormone activation-related dilutional effects.5–7
Hyperbilirubinemia, commonly noted in our sepsis group (17/31 [55%]) was also common in previous studies3–7 of cats with systemic infections and sepsis; icterus was also common in those studies. Sepsis-induced cholestasis has been described in people and dogs and is also manifested by high serum alkaline phosphatase and alanine transaminase activities56–58; however, its characteristic histopathologic findings were absent in cats with severe sepsis on necropsy.6 The low serum alkaline phosphatase activity in most (16/30 [53%]) and unremarkable serum γ-glutamyl-transferase activity in all cats of the sepsis group in our study were not supportive of cholestasis. Because alanine transaminase activity was typically within the RI, hepatocellular damage was not a likely cause of hyperbilirubinemia. Considering the frequency of anemia and the potential presence of low-grade hemolysis in cats with sepsis, hemolysis might be 1 cause of hyperbilirubinemia. Pancreatitis is a potential cause of icterus in cats59; however, specific markers such as 1,2-o-dilauryl-rac-glycero-3-glutaric acid-(6′-methylresorufin) ester lipase or feline pancreatic lipase immunoreactivity were not measured in this study, and this possibility warrants evaluation in future studies.
Hypertriglyceridemia has been detected in human patients with gram-negative bacterial infections and in laboratory animals experimentally infected with gram-negative bacteria or administered endotoxin,60–62 but it has not been previously reported in cats or dogs with sepsis. Nevertheless, hypertriglyceridemia was present in 22 of 30 (73%) cats with sepsis in the present study, although anorexia was common (18/30 [60%]). In people with sepsis, hypertriglyceridemia has been proposed to result from a high hepatic very-low-density lipoprotein secretion rate, a low very-low-density lipoprotein and triglyceride removal rate, or a combination of these factors,61 and its prognostic value in sepsis is debatable. Results of 1 study62 suggested that hypertriglyceridemia of sepsis may have a protective function, where lipoproteins bind endotoxin, preventing its interaction with lipopolysaccharide-binding proteins and uptake by macrophages, which triggers inflammatory mediator release. Conversely, in another human study,61 a circulating triglycerides concentration > 150 mg/dL at the onset of sepsis was a significant risk factor for death. In the present study, hypertriglyceridemia was not associated with death. Nevertheless, these findings were novel and suggest that circulating triglyceride concentrations should be further examined in cats with sepsis.
High serum CK and AST activities were common in cats with sepsis in our study, and these findings were consistent with muscle damage. This was possibly due to muscle ischemia, hypotension, shock, and hypoxia leading to rhabdomyolysis-like states in some cats, whereas in cats with bite wounds and cellulitis, direct muscle damage, trauma, and infection were likely more important causes. In human patients, rhabdomyolysis is biochemically characterized by myoglobinemia, myoglobinuria, and high serum muscle enzyme activities (in absence of myocardial injury), particularly CK activity ≥ 5 times the upper limit of the laboratory RI.63,64 The occurrence of rhabdomyolysis in cats of the present study could not be ascertained because serum or urine myoglobin concentrations were not measured. However, our findings were in agreement with a previous report65 indicating that sepsis was significantly more frequent among cats with CK activity above the RI than in cats in which CK activity was within the RI, and that markedly increased CK activity (median, 14,062 U/L; range, 278 to 97,829) was present in these patients, as well as in cats with systemic bacterial diseases.
The study reported here had several limitations. As previously mentioned, the sepsis group included only 31 cats, thereby limiting the statistical power. Additionally, the blood sample volume collected was insufficient for measuring all of the desired analytes in some cases because the health status of the cats was critical, and blood smears were lost or not evaluated for 5 patients. These missing data further limited some important statistical analyses (eg, lack of information on fibrinogen concentration, which also limited information on a diagnosis of DIC). The relatively low mortality rate (11/30 [37%]) also limited the ability to detect significant associations with the outcome of death. This relatively low mortality rate was likely partly attributable to case selection, with 10 of 31 (32%) cats having pyothorax, which is a septic condition but one that has a favorable prognosis.7,8 The heterogeneous underlying conditions in cats with sepsis also introduced variability; however, selecting a uniform population of cats with naturally occurring sepsis in a clinical setting would be difficult. Moreover, it might be argued that general conclusions regarding sepsis in cats could not be made if a uniform cohort population was selected.
The definition of sepsis used in the present study included evidence of SIRS with confirmed infection. In studies of sepsis in human medicine, specific definitions of sepsis, septic shock, and severe sepsis have been established to classify patients on the basis of inflammatory response severity.1,2 This allows for comparisons of results according to the severity of sepsis among various studies. However, there are currently no such consensus definitions for cats, and our study did not include use of any clinical illness severity scoring system that might have allowed assessing potential relationships between the degree of illness and laboratory findings.
Because established RIs for total protein C activity and antithrombin activity in cats were lacking, and since the study included only 33 healthy cats in the control group, which is an insufficient size for a reference population to establish RIs,66 the analytic procedures used were limited, and these values should not be extrapolated to other populations of cats. Future studies should therefore include a larger reference population, sufficient for establishing proper RIs for all hemostatic tests. In regard to laboratory methods, citrated plasma samples from cats with sepsis and healthy control cats were stored at −80°C for up to 12 months prior to analysis for hemostatic analytes. To the best of our knowledge, the stability of feline citrated plasma for such a period has not been examined. In previous studies23,67 of hemostasis in cats, plasma samples were stored for ≤ 6 months at various temperatures. A study68 of frozen human citrated plasma showed that storage at −74°C for ≥ 12 months resulted in variation of ≤ 5% for hemostatic analytes, and in a study69 of canine plasma, most hemostatic test results for samples frozen at −70°C for 6 months did not differ significantly from results for the same samples at the time of collection. Moreover, fresh frozen plasma stored at −18°C is considered stable, with full clotting factor activity for 1 year across species, including cats.70 We therefore believe that storage had little effect on our results; also, because healthy control cats were recruited during the same time period as ill cats and the samples were stored under identical conditions, any storage effect would have been equal between groups. We compared the results for cats with sepsis with those for healthy controls, and therefore, this potential limitation was considered minor.
Owing to the clinical nature of this study, laboratory findings were only examined at the time of initial examination at the study hospital, and serial measurements over the disease course would be needed to better understand the pathogenesis of sepsis, its progression, and trends over time in terms of associations with clinical changes or death. However, obtaining multiple blood samples of volumes sufficient for these tests in critically ill cats, especially in the face of anemia, would not have been ethical. Future studies should include larger, possibly more homogenous cohorts of cats, with collection of additional information (eg, arterial blood gas or oxygen saturation measurement, blood lactate concentration, and viscoelastic testing) and follow-up measurements where safely achievable. Additional information such as clinical severity score data and documentation of organ dysfunctions could help to improve the understanding of sepsis, to identify severely affected cats, and to define intervention points and treatment targets.
Acknowledgments
This manuscript includes a portion of a thesis submitted by Dr. Agi to the Koret School of Veterinary Medicine, Hebrew University of Jerusalem, as partial fulfillment of the requirements for a Doctor of Veterinary Medicine degree.
This study was supported by a grant from the Hebrew University Veterinary Teaching Hospital Clinical Study Fund.
The authors declare no conflicts of interest.
Presented in part in abstract form at the 20th International Veterinary Emergency and Critical Care Symposium, Indianapolis, Ind, September 2014.
ABBREVIATIONS
| aPTT | Activated partial thromboplastin time |
| AST | Aspartate transaminase |
| CK | Creatine kinase |
| DIC | Disseminated intravascular coagulopathy |
| PT | Prothrombin time |
| RI | Reference interval |
| SIRS | Systemic inflammatory response syndrome |
Footnotes
Advia 120, Siemens Medical Solutions Diagnostics, Erfurt, Germany.
Hematek slide strainer, Siemens Healthcare Diagnostics, Tarrytown, NY.
Cobas-Integra 400 Plus, Roche, Mannheim, Germany.
ACL-200 or ACL-9000 coagulometric analyzers, Instrumentation Laboratory, Milano, Italy.
HemosIL PT-fibrinogen recombinant 0020005000, Instrumentation Laboratory Co, Milano, Italy.
HemosIL AAPTT-SP liquid 0020006300, Instrumentation Laboratory Co, Milano, Italy.
HemosIL PT-fibrinogen 0008469810, Instrumentation Laboratory Co, Milano, Italy.
Chromogenic substrate, HemosIL 0020008910, Instrumentation Laboratory Co, Milano, Italy.
Factor Xa reagent, HemosIL 0020008920, Instrumentation Laboratory Co, Milano, Italy.
Protein C activator, HemosIL 0020300510, Instrumentation Laboratory Co, Milano, Italy.
Chromogenic substrate, HemosIL 0020300520, Instrumentation Laboratory Co, Milano, Italy.
Cobas-Integra 400 Plus, Roche, Mannheim, Germany.
Tina-quant D-dimer Gen 2, Roche, Mannheim, Germany.
SPSS, version 17.0 for Windows, SPSS Inc, Chicago, Ill.
Immuno Run Ag detection kit, Canine parvovirus, Biogal Galed Laboratory, Galed, Israel.
References
1. American College of Chest Physicians/Society of Critical Care Medicine consensus conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864–874.
2. Purvis D, Kirby R. Systemic inflammatory response syndrome: septic shock. Vet Clin North Am Small Anim Pract 1994; 24:1225–1247.
3. Declue AE, Delgado C, Chang CH, et al. Clinical and immunologic assessment of sepsis and the systemic inflammatory response syndrome in cats. J Am Vet Med Assoc 2011; 238:890–897.
4. Sergeeff JS, Armstrong PJ, Bunch SE. Hepatic abscesses in cats: 14 cases (1985–2002). J Vet Intern Med 2004; 18:295–300.
5. Costello MF, Drobatz KJ, Aronson LR, et al. Underlying cause, pathophysiologic abnormalities, and response to treatment in cats with septic peritonitis: 51 cases (1990–2001). J Am Vet Med Assoc 2004; 225:897–902.
6. Brady CA, Otto CM, Winkle TJ, et al. Severe sepsis in cats: 29 cases (1986–1998). J Am Vet Med Assoc 2000; 217:531–535.
7. Waddell LS, Brady CA, Drobatz KJ. Risk factors, prognostic indicators, and outcome of pyothorax in cats: 80 cases (1986–1999). J Am Vet Med Assoc 2002; 221:819–824.
8. Barrs VR, Allan GS, Martin P, et al. Feline pyothorax: a retrospective study of 27 cases in Australia. J Feline Med Surg 2005; 7:211–222.
9. Hopper K, Bateman S. An updated view of hemostasis: mechanisms of hemostatic dysfunction associated with sepsis. J Vet Emerg Crit Care 2005; 15:83–91.
10. Weiss DJ, Rashid J. The sepsis-coagulant axis: a review. J Vet Intern Med 1998; 12:317–324.
11. ten Cate H. Pathophysiology of disseminated intravascular coagulation in sepsis. Crit Care Med 2000; 28:S9–S11.
12. Angus DC, Van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013; 369:840–851.
13. Levi M, van der Poll T. Inflammation and coagulation. Crit Care Med 2010; 38:S26–S34.
14. Fourrier F. Severe sepsis, coagulation, and fibrinolysis: dead end or one way? Crit Care Med 2012; 40:2704–2708.
15. de Laforcade AM, Freeman LM, Shaw SP, et al. Hemostatic changes in dogs with naturally occurring sepsis. J Vet Intern Med 2003; 17:674–679.
16. de Laforcade AM, Rozanski EA, Freeman LM, et al. Serial evaluation of protein C and antithrombin in dogs with sepsis. J Vet Intern Med 2008; 22:26–30.
17. Bentley AM, Mayhew PD, Culp WTN, et al. Alterations in the hemostatic profiles of dogs with naturally occurring septic peritonitis. J Vet Emerg Crit Care (San Antonio) 2013; 23:14–22.
18. Declue AE. Sepsis and the systemic inflammatory response syndrome In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 7th ed. St Louis: Saunders-Elsevier, 2010; 523–528.
19. Weiss DJ. Uniform evaluation and semiquantitative reporting of hematologic data in veterinary laboratories. Vet Clin Pathol 1984; 13:27–31.
20. Stoackham SL, Scott MA. Leukocytes. In: Stoackham SL, Scott MA, eds. Fundamentals of veterinary clinical pathology. 2nd ed. Ames, Iowa: Iowa State Press, 2002; 76.
21. Weiss DJ, Souza CD. Monocytes and macrophages and their disorders. In: Weiss DJ, Wardrop KJ, eds. Schalm's veterinary hematology. 5th ed. Philadelphia: Wiley-Blackwell, 2000; 304.
22. Mackie IJ, Kitchen S, Machin SJ, et al. Guidelines on fibrinogen assays. Br J Haematol 2003; 121:396–404.
23. Dircks B, Nolte I, Mischke R. Haemostatic abnormalities in cats with naturally occurring liver diseases. Vet J 2012; 193:103–108.
24. Favaloro EJ. Laboratory testing in disseminated intravascular coagulation. Semin Thromb Hemost 2010; 36:458–467.
25. Wada H, Gabazza EC, Asakura H, et al. Comparison of diagnostic criteria for disseminated intravascular coagulation (DIC): diagnostic criteria of the International Society of Thrombosis and Hemostasis (ISTH) and of the Japanese Ministry of Health and Welfare for overt DIC. Am J Hematol 2003; 74:17–22.
26. Wada H, Asakura H, Okamoto K, et al. Expert consensus for the treatment of disseminated intravascular coagulation in Japan. Thromb Res 2010; 125:6–11.
27. Madden RM, Ward M, Marlar RA. Protein C activity levels in endotoxin-induced disseminated intravascular coagulation in a dog model. Thromb Res 1989; 55:297–307.
28. Boldt J, Papsdorf M, Rothe A, et al. Changes of the hemostatic network in critically ill patients—is there a difference between sepsis, trauma, and neurosurgery patients? Crit Care Med 2000; 28:445–450.
29. Lorente JA, Gardia-Frade LJ, Landin L, et al. Time course of hemostatic abnormalities in sepsis and its relation to outcome. Chest 1993; 103:1536–1542.
30. Hesselvik JF, Blomback M, Broden B, et al. Coagulation, fibrinolysis, and kallikrein systems in sepsis: relation to outcome. Crit Care Med 1989; 17:724–733.
31. Yan SB, Helterbrand JD, Hartman DL, et al. Low levels of protein C are associated with poor outcome in severe sepsis. Chest 2001; 120:915–922.
32. Goddard A, Wiinberg B, Schoeman JP, et al. Mortality in virulent canine babesiosis is associated with a consumptive coagulopathy. Vet J 2013; 196:213–217.
33. Estrin MA, Wehausen CE, Lessen CR, et al. Disseminated intravascular coagulation in cats. J Vet Intern Med 2006; 20:1334–1339.
34. Machida T, Kokubu H, Matsuda K, et al. Clinical use of D-dimer measurement for the diagnosis of disseminated intravascular coagulation in dogs. J Vet Med Sci 2010; 72:1301–1306.
35. Wiinberg B, Jensen AL, Johansson PI, et al. Development of a model based scoring system for diagnosis of canine disseminated intravascular coagulation with independent assessment of sensitivity and specificity. Vet J 2010; 185:292–298.
36. Esmon CT. Introduction: are natural anticoagulants candidates for modulating the inflammatory response to endotoxin? Blood 2000; 95:1113–1116.
37. White B, Perry D. Acquired antithrombin deficiency in sepsis. Br J Haematol 2001; 112:26–31.
38. Uchiba M, Okajima K, Murakami K, et al. Attenuation of endotoxin-induced pulmonary vascular injury by antithrombin III. Am J Physiol 1996; 270:L921–L930.
39. Ottenjann M, Weingart C, Arndt G, et al. Characterization of the anemia of inflammatory disease in cats with abscesses, pyothorax, or fat necrosis. J Vet Intern Med 2006; 20:1143–1150.
40. Weiss DJ, Krehbiel JD, Lund JE. Studies of the pathogenesis of anemia of inflammation: mechanism of impaired erythropoiesis. Am J Vet Res 1983; 44:1832–1835.
41. Weiss DJ, McClay CB. Studies on the pathogenesis of the erythrocyte destruction associated with the anemia of inflammatory disease. Vet Clin Pathol 1988; 17:90–93.
42. Nicolas G, Viatte L, Bennoun M, et al. Hepcidin, a new iron regulatory peptide. Blood Cells Mol Dis 2002; 29:327–335.
43. Babitt JL, Huang FW, Wrighting DM, et al. Bone morphogenetic protein signaling by hemojuvelin regulates hepcidin gene expression. Nat Genet 2006; 38:531–539.
44. Ben-Oz J, Segev G, Bilu G, et al. Peripheral nucleated red blood cells in cats and their association with age, laboratory findings, diseases, morbidity and mortality—a retrospective case-control study. Isr J Vet Med 2014; 69:181–190.
45. Brockus CW. Erythrocytes. In: Latimer KS, ed. Duncan and Prasse's veterinary laboratory medicine, clinical pathology. 5th ed. Ames, Iowa: Wiley–Blackwell, 2011; 3–44.
46. Declue AE, Williams KJ, Sharp C, et al. Systemic response to low-dose endotoxin infusion in cats. Vet Immunol Immunopathol 2009; 132:167–174.
47. Stachon A, Eisenblätter K, Köller B. Cytokines and erythropoietin in the blood of patients with erythroblasthemia. Acta Haematol 2003; 110:204–206.
48. Webb JL, Latimer KS. Leukocytes. In: Latimer KS, ed. Duncan and Prasse's veterinary laboratory medicine, clinical pathology. 5th ed. Ames, Iowa: Wiley-Blackwell, 2011; 45–82.
49. Watson PJ, Hertage ME. Hyperadrenocorticism in six cats. J Small Anim Pract 1998; 39:175–184.
50. Burton S. Clinical pathology interpretation. Can Vet J 1997; 38:179–181.
51. Burton AG, Harris LA, Owens SD, et al. Degenerative left shift as a prognostic tool in cats. J Vet Intern Med 2014; 28:912–917.
52. Nivy R, Itkin Y, Bdolah-Abram T, et al. Neutrophil counts and morphology in cats: a retrospective case-control study of 517 cases. Isr J Vet Med 2013; 68:149–157.
53. de Morais H, Boiondo A. Disorders of chloride: hyperchloremia and hypochloremia. In: Dibartola S, ed. Fluid, electrolyte and acid-base disorders in small animal practice. 3rd ed. St Louis: Saunders-Elsevier, 2006; 80–91.
54. Kellett Gregory LM, Mittleman-Boller E, Brown DC, et al. Ionized calcium concentrations in cats with septic peritonitis: 55 cases (1990–2008). J Vet Emerg Crit Care (San Antonio) 2010; 20:398–405.
55. Bourke E, Delaney V. Assessment of hypocalcemia and hypercalcemia. Clin Lab Med 1993; 13:157–181.
56. Taboada J, Meyer D. Cholestasis associated with extrahepatic bacterial infection in five dogs. J Vet Intern Med 1989; 3:216–221.
57. Chand N, Sanyal AJ. Sepsis-induced cholestasis. Hepatology 2007; 45:230–241.
58. Moseley RH. Sepsis and cholestasis. Clin Liver Dis 2004; 8:83–94.
59. Ferreri JA, Hardam E, Kimmel SE, et al. Clinical differentiation of acute necrotizing from chronic nonsuppurative pancreatitis in cats: 63 cases (1996–2001). J Am Vet Med Assoc 2003; 223:469–474.
60. Scholl RA, Lang CH, Bagby GJ. Hypertriglyceridemia and its relation to tissue lipoprotein lipase activity in endotoxemic, Escherichia coli bacteremic and polymicrobial septic rats. J Surg Res 1984; 37:394–401.
61. Cetinkaya A, Erden A, Avci D, et al. Is hypertriglyceridemia a prognostic factor in sepsis? Ther Clin Risk Manag 2014; 10:147–150.
62. Harris HW, Grunfeld C, Feingold KR, et al. Human very low density lipoproteins and chylomicrons can protect against endotoxin-induced death in mice. J Clin Invest 1990; 86:696–702.
63. Melli G, Chaudhry V, Cornblath DR. Rhabdomyolysis: an evaluation of 475 hospitalized patients. Medicine (Baltimore) 2005; 84:377–385.
64. Poels PJ, Gabreels FJ. Rhabdomyolysis: a review of the literature. Clin Neurol Neurosurg 1993; 95:175–192.
65. Aroch I, Keidar I, Himmelstein A, et al. Diagnostic and prognostic value of serum creatine-kinase activity in ill cats: a retrospective study of 601 cases. J Feline Med Surg 2010; 12:466–475.
66. Friedrichs KR, Harr KE, Freeman KP, et al. ASVCP reference interval guidelines: determination of de novo reference intervals in veterinary species and other related topics. Vet Clin Pathol 2012; 41:441–453.
67. Brazzell JL, Borjesson DL. Evaluation of plasma antithrombin activity and D-dimer concentration in populations of healthy cats, clinically ill cats, and cats with cardiomyopathy. Vet Clin Pathol 2007; 36:79–84.
68. Woodhams B, Girardot O, Blanco MJ, et al. Stability of coagulation proteins in frozen plasma. Blood Coagul Fibrinolysis 2001; 12:229–236.
69. Bateman SW, Mathews KA, Abrams-Ogg ACG, et al. Evaluation of the effect of storage at 70 degrees C for six months on hemostatic function testing in dogs. Can J Vet Res 1999; 63:216–220.
70. Brecher ME. AABB technical manual. 15th ed. Bethesda, Md: American Association of Blood Banks; 2005.
