Abstract
OBJECTIVE To determine the incidence of blood transfusion, mortality rate, and factors associated with transfusion in dogs and cats undergoing liver lobectomy.
DESIGN Retrospective case series.
ANIMALS 63 client-owned dogs and 9-client owned cats that underwent liver lobectomy at a specialty veterinary practice from August 2007 through June 2015.
PROCEDURES Medical records were reviewed and data extracted regarding dog and cat signalment, hematologic test results before and after surgery, surgical method, number and identity of lobes removed, concurrent surgical procedures, hemoabdomen detected during surgery, incidence of blood transfusion, and survival to hospital discharge (for calculation of mortality rate). Variables were compared between patients that did and did not require transfusion.
RESULTS 11 of 63 (17%) dogs and 4 of 9 cats required a blood transfusion. Mortality rate was 8% for dogs and 22% for cats. Pre- and postoperative PCV and plasma total solids concentration were significantly lower and mortality rate significantly higher in dogs requiring transfusion than in dogs not requiring transfusion. Postoperative PCV was significantly lower in cats requiring transfusion than in cats not requiring transfusion. No significant differences in any other variable were identified between dogs and cats requiring versus not requiring transfusion.
CONCLUSIONS AND CLINICAL RELEVANCE Dogs and cats undergoing liver lobectomy had a high requirement for blood transfusion, and a higher requirement for transfusion should be anticipated in dogs with perioperative anemia and cats with postoperative anemia. Veterinarians performing liver lobectomies in dogs and cats should have blood products readily available.
Liver lobectomies are performed in dogs to treat hepatic neoplasia, trauma, or torsion.1–4 Primary liver tumors constitute only 0.6% to 1.3% of neoplasms in dogs, with hepatocellular carcinoma representing the most common type, although bile duct carcinoma, carcinoids, and sarcoma have also been identified.2 Liver lobe torsion is uncommon and most frequently involves the left lateral liver lobe in dogs, with surgery indicated for repositioning or excision of the affected lobe.4,5
In cats, hepatic masses and liver lobe torsion are also indications for liver lobectomy.4,6 Primary hepatic tumors account for 1% to 3% of neoplasms in cats, with hepatocellular adenoma more common than hepatocellular carcinoma. Biliary cystadenomas account for > 50% of all hepatobiliary tumors, whereas bile duct carcinoma is the most common malignant hepatobiliary tumor in cats.6
Techniques for liver lobectomy in dogs and cats include dissection and ligation, placement of an encircling ligature around the base of the lobe, use of an electrosurgical vessel sealing device, or use of a thoracoabdominal stapler.1 The left hepatic lobes are more easily removed with any technique than other lobes because the hilus is more easily accessed.7 Dissection and ligation involve blunt dissection of the liver parenchyma at the hilus, allowing for ligation of individual ducts and vessels and thus decreasing the potential for ligature slippage and hemorrhage.1 The encircling ligature technique can lead to an increased risk of hemorrhage and bile leakage when used in large dogs or for lobes other than the left medial and left lateral lobes; therefore, its use for resection of these lobes alone is only recommended for small patients.1 Use of a thoracoabdominal stapler is reportedly a consistent and efficient technique for liver lobectomy, offering the advantages of speed, more complete excision, and reduced inflammation at the excision site, compared with other techniques for lobectomy.1,8
Hemorrhage is a commonly observed surgical complication following liver lobectomy in dogs and cats. In a case series9 of 48 dogs undergoing complete liver lobectomy for massive hepatocellular carcinoma, major intraoperative hemorrhage occurred in 7% of patients, moderate hemorrhage in 2%, and minor hemorrhage in 14%. Two of the 48 (4%) dogs died of intraoperative exsanguination.9 In another case series10 involving 29 patients of various species, including cats and dogs, mild intraoperative bleeding was noted during 4 of 32 (12%) lobectomies, and 1 dog developed hemoabdomen after lobectomy and received a packed RBC transfusion. Six of 15 dogs undergoing liver lobectomy received a blood transfusion in a third case series.11 In a case series12 involving 18 cats undergoing liver lobectomy, transfusion was reportedly required for 11 (61%) cats.12 Hemorrhage can also occur during liver biopsy, with 4% of small animal patients undergoing laparoscopic liver biopsy reportedly requiring transfusion.13 These reports suggest that hemorrhage requiring blood transfusion is often encountered on an urgent basis during liver lobectomy in both dogs and cats.
The purpose of the study reported here was to determine the incidence of blood transfusion and postoperative mortality rate (based on survival to hospital discharge) in dogs and cats undergoing liver lobectomy. A secondary objective was to evaluate potential associations between specific factors (age, body weight, preoperative hemoabdomen, pre- and postoperative PCV and plasma total solids concentration, surgical method of lobectomy, number and specific lobe or lobes removed, concurrent performance of other surgical procedures, and survival to hospital discharge) and the need for perioperative blood transfusion. We hypothesized that liver lobectomy would be associated with a high incidence of blood transfusion but a low mortality rate in dogs and cats. In addition, we hypothesized that there would be an association between the secondary factors evaluated and need for perioperative blood transfusion.
Materials and Methods
Case selection criteria
Electronic medical records of a specialty veterinary practice from August 2007 through June 2015 were searched to identify dogs and cats that underwent liver lobectomy during this period. To be included in the study, dogs and cats were required to have a complete surgical report and medical records detailing pre- and postoperative care, monitoring data, and outcome.
Medical records review
Medical records were reviewed and data extracted regarding dog and cat signalment, hematologic test results (PCV and plasma total solids concentration) before and ≤ 1 hour after surgery, surgical method, number and identity of lobes removed, concurrent surgical procedures, hemoabdomen detected during surgery, whether and when a blood transfusion was performed, and survival to hospital discharge (for calculation of mortality rate). Administration of blood products and timing relative to surgery were also noted. Blood, packed RBCs, or an autotransfusion was administered on the basis of product availability, and these products were obtained from either a national blood bank or from the hospital's in-house blood donor program. All surgical procedures were performed by a diplomate of the American College of Veterinary Surgeons.
Statistical analysis
All data were analyzed by use of statistical software.a Incidence of blood transfusion at various points relative to surgery and mortality rate were calculated for both dogs and cats. Collected patient and surgical data were compared between dogs requiring or not requiring a blood transfusion. For cats, only data regarding age, body weight, and pre- and postoperative PCV and plasma total solids concentration were compared between those requiring or not requiring transfusion because of the small sample size. Continuous data were tested for normality of distribution with the Shapiro-Wilk test. The paired t test was used to compare means for normally distributed data, whereas the Mann-Whitney U test was used to compare nonnormally distributed data. For categorical data comparisons, the χ2 test was used. Values of P ≤ 0.05 were considered significant.
Results
Seventy-six dogs (n = 67) and cats (9) underwent liver lobectomy from August 2007 through June 2015 and were identified as eligible for the study. Four dogs were excluded because of a lack of recorded pre- or postoperative measurement of PCV or plasma total solids concentration, leaving 63 dogs in the study.
Dogs
Mean ± SD age of the 63 dogs included in the study was 10.7 ± 2.0 years, and mean body weight was 25.2 ± 12.5 kg (55.4 ± 27.5 lb). Forty (63%) dogs were spayed females, 21 (33%) were castrated males, 1 (2%) was a sexually intact female, and 1 (2%) was a sexually intact male. Dogs were classified as Labrador Retriever (n = 7), Golden Retriever (7), German Shepherd Dog (6), mixed-breed dog (4), Airedale Terrier (4), Shih Tzu (3), West Highland White Terrier (3), Australian Cattle Dog (2), German Shorthair Pointer (2), Golden Retriever–Poodle mix (2), Rhodesian Ridgeback (2), Shetland Sheepdog (2), Siberian Husky (2), Yorkshire Terrier (1), and 1 each of 15 other breeds.
Eleven (17%) dogs required a blood transfusion, and 52 (83%) did not. Three dogs received the transfusion during surgery, and 8 dogs received it after surgery. Of the 3 dogs that received the transfusion during surgery, 2 received a packed RBC transfusion and 1 received an autotransfusion. Four of the dogs that received a transfusion after surgery received packed RBCs and 4 received whole blood. Mean ± SD age of dogs requiring a blood transfusion was 11.2 ± 1.5 years and did not differ significantly (P = 0.34) from that of dogs requiring no transfusion (10.6 ± 2.0 years). Mean ± SD body weight of dogs requiring a blood transfusion was 24.6 ± 12.2 kg (54.1 ± 26.8 lb) and did not differ significantly (P = 0.84) from that of dogs not requiring a transfusion (25.4 ± 12.7 kg [55.9 ± 27.9 lb]).
Significant differences in pre- and postoperative (≤ 1 hour after surgery) PCV and plasma total solids concentration were identified between dogs requiring versus not requiring a transfusion (Table 1). No such differences were identified between dogs that received the transfusion during surgery and dogs that received it after surgery (Table 2).
Mean ± SD pre- and postoperative (≤ 1 hour after surgery) PCV and plasma total solids concentration in client-owned dogs undergoing liver lobectomy that required (n = 11) or did not require (52) a blood transfusion.
| Variable | Transfusion | No transfusion | P value |
|---|---|---|---|
| Preoperative PCV (%) | 31 ± 7 | 40 ± 8 | 0.002 |
| Preoperative total solids (g/dL) | 6.3 ± 1.6 | 7.4 ± 1.2 | 0.02 |
| Postoperative PCV (%) | 24 ± 9 | 35 ± 9 | 0.004 |
| Postoperative total solids (g/dL) | 4.7 ± 0.8 | 6.0 ± 1.3 | 0.001 |
Values of P ≤ 0.05 were considered significant (Mann-Whitney U test).
Mean ± SD pre- and postoperative (≤ 1 hour after surgery) PCV and plasma total solids concentration in client-owned dogs undergoing liver lobectomy that required a blood transfusion during surgery (n = 3) or after surgery (8).
| Variable | Intraoperative | Postoperative | P value |
|---|---|---|---|
| Preoperative PCV (%) | 29 ± 10 | 32 ± 6 | 0.63 |
| Preoperative total solids (g/dL) | 6.3 ± 0.9 | 6.3 ± 1.9 | 0.88 |
| Postoperative PCV (%) | 30 ± 1 | 22 ± 9 | 0.06 |
| Postoperative total solids (g/dL) | 5.2 ± 0.3 | 4.5 ± 0.8 | 0.14 |
Values of P ≤ 0.05 were considered significant (Mann-Whitney U test).
Forty-six (73%) dogs underwent a single liver lobectomy, 16 (25%) dogs had 2 lobes removed, and 1 (2%) dog had 3 lobes removed. Neither the lobe removed nor total number of lobes removed differed significantly between dogs that received and did not receive a blood transfusion (Table 3). Surgical technique used for liver lobectomy was summarized (Table 4), revealing no significant differences in techniques used between dogs requiring versus not requiring a transfusion.
Number of dogs in Table 1 with various liver lobes removed, categorized by whether they required (n = 11) or did not require (52) a blood transfusion.
| Lobe removed | Transfusion | No transfusion | P value |
|---|---|---|---|
| Left lateral | 6 | 22 | 0.74 |
| Left medial | 4 | 14 | 0.80 |
| Quadrate | 1 | 15 | 0.30 |
| Caudate | 1 | 6 | 0.66 |
| Right medial | 1 | 5 | 0.62 |
| Right lateral | 1 | 5 | 0.12 |
Values of P ≤ 0.05 were considered significant (χ2 test).
Number of dogs in Table 1 in which various surgical techniques were performed for liver lobectomy, categorized by whether they required (n = 11) or did not require (52) a blood transfusion.
| Surgical technique | Transfusion | No transfusion | P value |
|---|---|---|---|
| Surgical stapling deviceb | 9 | 32 | 0.20 |
| Combination of techniques | 1 | 9 | 0.50 |
| Electrosurgical vessel sealing devicec | 1 | 7 | 0.69 |
| Ligation | 0 | 2 | 0.51 |
| Not recorded | 0 | 2 | 0.51 |
Values of P ≤ 0.05 were considered significant (χ2 test).
Hemoabdomen was identified at the time of surgery in 11 of the 63 (17%) dogs. Three of the dogs with hemoabdomen received a blood transfusion, compared with 8 dogs without hemoabdomen that received a transfusion; the difference was not significant (P = 0.34). Concurrent surgical procedures in addition to liver lobectomy were performed in 34 (54%) dogs and included splenectomy (n = 11), collection of intestinal biopsy specimens (8), cholecystectomy (3), cystotomy (3), diaphragmatic hernia repair (2), and various other procedures. No significant (P = 0.39) difference was identified between dogs requiring versus not requiring a transfusion with regard to having concurrent procedures performed or the number of concurrent procedures performed.
Five (8%) dogs undergoing liver lobectomy did not survive to hospital discharge. Three of the 11 (27%) dogs requiring a blood transfusion did not survive to hospital discharge, compared with 2 of the 52 (4%) dogs not requiring transfusion; this difference was significant (P = 0.01).
Cats
Mean ± SD age of the 9 included cats was 15.1 ± 3.2 years, and mean body weight was 4.1 ± 1.3 kg (9.0 ± 2.9 lb). Five cats were castrated males, and 4 were spayed females. Seven cats were domestic shorthairs, and 2 were Himalayans.
Four cats required a blood transfusion. Mean ± SD age of cats that received a transfusion was 14.5 ± 4.8 years and did not differ significantly (P = 0.74) from that of cats requiring no transfusion (15.6 ± 1.7 years). Mean ± SD body weight of cats requiring a transfusion was 4.3 ± 1.9 kg (9.5 ± 4.2 lb) and also did not differ significantly (P = 0.28) from that of cats requiring no transfusion (3.9 ± 0.6 kg [8.6 ± 1.3 lb]).
One of the 4 cats requiring a transfusion received it during surgery, with the blood product administered at the time of cardiac arrest in conjunction with CPR. Return of spontaneous circulation was not achieved, and neither the cause of arrest nor the rationale for transfusion in this patient was documented in the medical record. The remaining 3 cats received a transfusion after surgery.
No significant difference was identified in preoperative PCV or plasma total solids concentration between cats requiring versus not requiring a blood transfusion (Table 5). Packed cell volume ≤ 1 hour after surgery, but not plasma total solids concentration at that point, was significantly lower in cats requiring versus not requiring a transfusion
Mean ± SD pre- and postoperative (≤ 1 hour after surgery) PCV and plasma total solids concentration in client-owned cats undergoing liver lobectomy that required (n = 4) or did not require (5) a blood transfusion.
| Variable | Transfusion | No transfusion | P value |
|---|---|---|---|
| Preoperative PCV (%) | 25 ± 10 | 33 ± 4 | 0.14 |
| Preoperative total solids (g/dL) | 6.6 ± 2.1 | 7.7 ± 0.7 | 0.71 |
| Postoperative PCV (%)* | 14 ± 2 | 25 ± 8 | 0.04 |
| Postoperative total solids (g/dL)* | 4.0 ± 1.1 | 5.5 ± 0.4 | 0.15 |
In the transfusion group, data represent 3 cats; 1 cat died while anesthetized and receiving an intraoperative transfusion.
Values of P ≤ 0.05 were considered significant (Mann-Whitney U test).
Four cats underwent left lateral lobectomy, and 1 cat had the quadrate lobe removed. Two cats had both the left lateral and left medial lobes removed, 1 had the left and right medial lobes removed, and 1 had the right medial and quadrate lobes removed. The left lateral lobe was the lobe most commonly removed, accounting for 6 of 12 excised lobes. A thoracoabdominal stapling deviceb was used for liver lobectomy in 5 cats, an electrosurgical vessel sealing devicec in 3 cats, and an encircling ligature in 1 cat.
No hemoabdomen was identified in any cat at the time of liver lobectomy. Five of the 9 cats underwent concurrent surgical procedures, including collection of intestinal biopsy specimens (n = 2), intestinal biopsy and resection anastomosis (1), removal of a body wall mass (1), and complete right mastectomy (1).
Two of the 9 (22%) cats failed to survive to hospital discharge, including a cat that developed cardiac arrest while anesthetized and receiving a blood transfusion and another cat that became progressively anemic during the 24-hour period following surgery, prompting the owners to elect euthanasia without attempting transfusion. A proposed source of anemia was not discussed in the medical record of this patient.
Discussion
In the study reported here, 17% of dogs undergoing liver lobectomy required a blood transfusion, which was lower than the percentage of dogs in a previous study,11 in which 6 of 15 (40%) dogs undergoing liver lobectomy required a blood transfusion. Although an earlier study9 revealed a high rate of hemorrhage in dogs undergoing liver lobectomy (10/48 [21%]), perioperative transfusion administration was not assessed. Differences among studies in the need for transfusion might be attributable to differences in patient populations, surgeon skill, or criteria used for transfusion administration or to small sample sizes. Ultimately, the retrospective nature of these studies makes it difficult to determine the exact reason for any differences observed.
The overall mortality rate for dogs in the present study was 8%, which is consistent with previously reported mortality rates.9,11 A higher mortality rate was observed in dogs requiring transfusion, with 3 of 11 (27%) dogs failing to survive to hospital discharge. The higher mortality rate for dogs requiring versus not requiring a transfusion could have been the result of greater disease severity, but specific injury and disease severity scores could not be assessed because of limited data availability.
A significant difference was identified in pre- and postoperative (≤ 1 hour after surgery) PCV and plasma total solids concentration, with dogs requiring a transfusion at any point (during or after surgery) significantly more likely to be anemic before surgery than dogs not requiring transfusion. Because of the retrospective nature of the present study, the reason for transfusion administration was not recorded for most patients and was at the discretion of the attending clinician. However, given our findings, close perioperative monitoring of PCV and plasma total solids concentration is recommended for dogs undergoing liver lobectomy.
Approximately one-quarter (3/11) of dogs with hemoabdomen in the present study required blood transfusion, which is an incidence that warrants ensuring ready access to blood products when treating similar dogs. In a previous retrospective study14 of 83 dogs treated surgically for hemoabdomen secondary to various underlying etiologies, 11 underwent liver lobectomy. All dogs in that study required blood transfusion regardless of the origin of the hemoabdomen. Prospective evaluation of dogs with hemoabdomen undergoing liver lobectomy would be valuable, including examination of the volume of abdominal effusion present, hematologic variables, and the need for transfusion support, to determine whether dogs with hemoabdomen are indeed more likely to require transfusion than dogs without hemoabdomen.
An association between body weight and the need for perioperative RBC transfusion has been identified in dogs undergoing various surgical procedures.11 In the present study, however, no significant difference in body weight was identified between dogs requiring versus not requiring a blood transfusion. This difference in findings cannot be immediately explained, particularly given that mean ± SD body weights (25.2 ± 12.5 kg in the present study and 23.3 ± 3.2 kg [51.3 ± 7.0 lb] for dogs requiring liver lobectomy11) were similar between studies, and may have been attributable to different sample sizes or specific surgical procedures.
In comparisons among surgical techniques for liver lobectomy in the present study, no significant differences were identified in the proportion of dogs requiring transfusion. However, the study was performed at a single institution, where most dogs underwent liver lobectomy with a surgical stapling device. Because of the small number of dogs in which other techniques were used, which limited statistical power to detect significant differences, further investigation into the relationship between liver lobectomy technique and blood transfusion requirement is needed before it can be concluded that technique has no effect.
Four of 9 (44%) cats received a blood transfusion in the study reported here, which was less than a previously reported rate of 61% in cats.12 A small sample size and differences in patient populations and surgical techniques may underlie this difference. No significant difference in preoperative PCV or plasma total solids concentration was identified between cats requiring versus not requiring a blood transfusion; however, the small sample size likely contributed to low statistical power. Cats requiring a blood transfusion after liver lobectomy had a significantly lower postoperative PCV than did those not requiring transfusion, indicating that postoperative PCV should be monitored closely in cats after liver lobectomy regardless of hematologic findings prior to surgery. Cats in the present study had a high perioperative mortality rate of 22%, which was identical to the rate identified in another study.12 Although the small sample size may have influenced the mortality rate and the incidence of transfusion requirement, both findings warrant serious attention when planning for liver lobectomy in cats.
The present study had several limitations attributable to its retrospective nature, including lack of disease severity scores for comparison between groups, inability to determine the definitive reason for transfusion administration in some patients, and a small number of included cats. Calculation of disease severity scores would allow for a more objective comparison of groups at risk of requiring blood transfusion to identify specific subgroups at greater risk than others. Prospective evaluation with emphasis on identification of the particular reasons for transfusion administration in dogs and cats undergoing liver lobectomy may help in the determination of specific indications for transfusion.
Regardless of the aforementioned limitations, findings of the present study suggested that a clinically important percentage of dogs and cats undergoing liver lobectomy required blood transfusion and the mortality rate associated with liver lobectomy was high, particularly for cats. The higher incidence of transfusion requirement in dogs undergoing liver lobectomy with hemoabdomen should be evaluated in a larger prospective study. Presurgical planning for both dogs and cats undergoing liver lobectomy should include discussion with the owners regarding the possible need for blood transfusion. Additional preparation should include ensuring access to blood products for use in the perioperative period, and blood typing and cross-matching may also be advisable.
Footnotes
Stata, version 13, StataCorp LP, College Station, Tex.
TA thoraco-abdominal stapler, Covidien Animal Health, Mansfield, Mass.
LigaSure vessel sealing device, Covidien Animal Health, Mansfield, Mass.
References
1. May LR, Mehler SJ. Complications of hepatic surgery in companion animals. Vet Clin North Am Small Anim Pract 2011; 41: 935–948.
2. Mayhew PD, Weisse C. Liver and biliary system. In: Tobias KM, Johnston SA, eds. Veterinary surgery: small animal. St Louis: Saunders, an imprint of Elsevier, 2012; 1601–1623.
3. Scherk MA, Center SA. Toxic, metabolic, infectious and neoplastic liver diseases. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 7th ed. St Louis: Saunders, an imprint of Elsevier, 2010; 1672–1689.
4. Swann HM, Brown DC. Hepatic lobe torsion in 3 dogs and a cat. Vet Surg 2001; 30: 482–486.
5. Schwartz SG, Mitchell SL, Keating JH, et al. Liver lobe torsion in dogs: 13 cases (1995–2004). J Am Vet Med Assoc 2006; 228: 242–247.
6. Liptak JM, Dernell WS, Withrow SJ. Liver tumors in dogs and cats. Compend Contin Educ Pract Vet 2004; 26: 50–57.
7. Boothe HW. Current concepts in hepatobiliary surgery. Vet Clin North Am Small Anim Pract 2015; 45: 463–475.
8. Lewis DD, Bellenger CR, Lewis DT, et al. Hepatic lobectomy in the dog: a comparison of stapling and ligation techniques. Vet Surg 1990; 19: 221–225.
9. Liptak JM, Dernell WS, Monnet E, et al. Massive hepatocellular carcinoma in dogs: 48 cases (1992–2002). J Am Vet Med Assoc 2004; 225: 1225–1230.
10. Goodman AR, Casale SA. Short-term outcome following partial or complete liver lobectomy with a commercially prepared self-ligating loop in companion animals: 29 cases (2009–2012). J Am Vet Med Assoc 2014; 244: 693–698.
11. Haley AL, Mann FA, Middleton J, et al. Perioperative red blood cell transfusion requirement for various surgical procedures in dogs: 207 cases (2004–2013). J Am Vet Med Assoc 2015; 247: 85–91.
12. Pavia PR, Kovak-McClaran J, Lamb K. Outcome following liver lobectomy using thoracoabdominal staplers in cats. J Small Anim Pract 2014; 55: 22–27.
13. McClaran JK, Buote NJ. Complications and need for conversion to laparotomy in small animals. Vet Clin North Am Small Anim Pract 2009; 39: 941–951.
14. Lux CN, Culp WT, Mayhew PD, et al. Perioperative outcome in dogs with hemoperitoneum: 83 cases (2005–2010). J Am Vet Med Assoc 2013; 242: 1385–1391.
