Evaluation of plasma lactate concentration and base excess at the time of hospital admission as predictors of gastric necrosis and outcome and correlation between those variables in dogs with gastric dilatation-volvulus: 78 cases (2004–2009)

Kari A. Santoro Beer Section of Critical Care, Department of Clinical Studies, Matthew J. Ryan Veterinary Hospital, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104.

Search for other papers by Kari A. Santoro Beer in
Current site
Google Scholar
PubMed
Close
 DVM
,
Rebecca S. Syring Section of Critical Care, Department of Clinical Studies, Matthew J. Ryan Veterinary Hospital, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104.

Search for other papers by Rebecca S. Syring in
Current site
Google Scholar
PubMed
Close
 DVM, DACVECC
, and
Kenneth J. Drobatz Section of Critical Care, Department of Clinical Studies, Matthew J. Ryan Veterinary Hospital, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104.

Search for other papers by Kenneth J. Drobatz in
Current site
Google Scholar
PubMed
Close
 DVM, MSCE, DACVECC, DACVIM

Click on author name to view affiliation information

Abstract

Objective—To determine the correlation between plasma lactate concentration and base excess at the time of hospital admission and evaluate each variable as a predictor of gastric necrosis or outcome in dogs with gastric dilatation-volvulus (GDV).

Design—Retrospective case series.

Animals—78 dogs.

Procedures—For each dog, various data, including plasma lactate concentration and base excess at the time of hospital admission, surgical or necropsy findings, and outcome, were collected from medical records.

Results—Gastric necrosis was identified in 12 dogs at the time of surgery and in 4 dogs at necropsy. Sixty-five (83%) dogs survived to hospital discharge, whereas 13 (17%) dogs died or were euthanized. Of the 65 survivors and 8 nonsurvivors that underwent surgery, gastric necrosis was detected in 8 and 4 dogs, respectively. Via receiver operating characteristic curve analysis, an initial plasma lactate concentration cutoff of 7.4 mmol/L was 82% accurate for predicting gastric necrosis (sensitivity, 50%; specificity, 88%) and 88% accurate for predicting outcome (sensitivity, 75%; specificity, 89%). Among all dogs, the correlation between initial plasma lactate concentration and base excess was significant, although base excess was a poor discriminator for predicting gastric necrosis or outcome (area under the receiver operating characteristic curve, 0.571 and 0.565, respectively).

Conclusions and Clinical Relevance—In dogs with GDV, plasma lactate concentration at the time of hospital admission was a good predictor of gastric necrosis and outcome. However, despite the correlation between initial base excess and plasma lactate concentration, base excess should not be used for prediction of gastric necrosis or outcome in those patients.

Abstract

Objective—To determine the correlation between plasma lactate concentration and base excess at the time of hospital admission and evaluate each variable as a predictor of gastric necrosis or outcome in dogs with gastric dilatation-volvulus (GDV).

Design—Retrospective case series.

Animals—78 dogs.

Procedures—For each dog, various data, including plasma lactate concentration and base excess at the time of hospital admission, surgical or necropsy findings, and outcome, were collected from medical records.

Results—Gastric necrosis was identified in 12 dogs at the time of surgery and in 4 dogs at necropsy. Sixty-five (83%) dogs survived to hospital discharge, whereas 13 (17%) dogs died or were euthanized. Of the 65 survivors and 8 nonsurvivors that underwent surgery, gastric necrosis was detected in 8 and 4 dogs, respectively. Via receiver operating characteristic curve analysis, an initial plasma lactate concentration cutoff of 7.4 mmol/L was 82% accurate for predicting gastric necrosis (sensitivity, 50%; specificity, 88%) and 88% accurate for predicting outcome (sensitivity, 75%; specificity, 89%). Among all dogs, the correlation between initial plasma lactate concentration and base excess was significant, although base excess was a poor discriminator for predicting gastric necrosis or outcome (area under the receiver operating characteristic curve, 0.571 and 0.565, respectively).

Conclusions and Clinical Relevance—In dogs with GDV, plasma lactate concentration at the time of hospital admission was a good predictor of gastric necrosis and outcome. However, despite the correlation between initial base excess and plasma lactate concentration, base excess should not be used for prediction of gastric necrosis or outcome in those patients.

Gastric dilatation-volvulus is a life-threatening condition in dogs that requires rapid surgical intervention, with reported mortality rates ranging from 13% to 43%.1–5 Affected dogs often have inadequate tissue perfusion as a result of decreased venous return caused by gastric distention and obstruction of the caudal vena cava. This hypoperfusion of tissue beds leads to anaerobic metabolism, frequently inducing metabolic acidosis as a result of hyperlactatemia.6 Gastric necrosis, an additional sequela of hypoperfusion of the gastric mucosa, has been reported as an important predictor of outcome among dogs with GDV.1–3,7 Results of studies1,8 have indicated that the degree of hyperlactatemia offers prognostic information for these patients and may also be used to predict the likelihood of gastric necrosis.

Base excess is defined as the amount of strong acid or base required to restore 1 L of blood to a pH of 7.40 at 37°C with Pco2 held constant at 40 mm Hg.9 Base excess is a calculated value derived from the Siggaard-Andersen nomogram and is dependent on blood pH, Pco2, and Hct. As the circulating concentration of lactate (a strong acid) increases, base excess should proportionately decrease. Current thinking is that there is a nearly stoichiometric relationship between base excess and lactate concentration in blood.10,11

In species used in experimental models of various disease processes, a correlation between blood lactate concentration and base excess has been identified, with initial values providing prognostic information in some instances.12–14 Resolution of abnormalities in those 2 variables is frequently used as an endpoint of resuscitation efforts. In 1991, Dunham et al14 reported that base excess correlated most closely with developing oxygen debt and subsequent death in dogs with experimentally induced hemorrhagic shock; however, the combination of blood lactate concentration and base excess assessments was superior in that regard to either assessment alone. The predictive ability of base excess with regard to survival rate was confirmed in an investigation of humans with blunt trauma; results indicated that base excess at the time of admission to hospital was the single most important variable in predicting probability of death, although blood lactate concentration at the time of admission to hospital also had a high predictive value.15 In pigs with experimentally induced hemorrhagic shock, Davis et al13 found that base excess not only was correlated with the hemodynamic and tissue perfusion changes associated with shock but also was correlated with those changes during resuscitation. In a subsequent study,12 it was determined that base excess correlated well with serum lactate concentration. Although base excess and circulating lactate concentration each correlated well to shock and resuscitation in those studies,12,13 other investigations16,17 in humans revealed poor correlation between serum lactate concentration and base excess either initially or after resuscitation.

To the authors' knowledge, the correlation between initial plasma lactate concentration and base excess has not been investigated in clinical veterinary medicine. The purpose of the study reported here was to determine the correlation between plasma lactate concentration and base excess at the time of hospital admission and evaluate each variable as a predictor of gastric necrosis or outcome in dogs with GDV.

Materials and Methods

Case selection—Medical records of all dogs examined at the University of Pennsylvania Matthew J. Ryan Veterinary Hospital for GDV between January 2004 and July 2009 were reviewed. Dogs with a radiographic diagnosis of GDV and results of venous blood gas evaluation performed at the time of admission to the hospital were included in the study. Dogs that received fluid therapy or gastric decompression prior to initial evaluation, had an incomplete medical record or lack of initial venous blood gas analysis data, or were euthanized prior to surgery and did not undergo necropsy to assess gastric viability were excluded from the study. Dogs with a concurrent disease process that might also induce metabolic acidosis or affect lactate production or clearance (eg, uremia, exogenous acid administration or toxicosis, diabetic ketoacidosis, neoplasia, or liver failure) were also excluded.

Medical records review—Information regarding signalment, history, fluid therapy, initial plasma lactate concentration and base excess data, duration of hospitalization (including time spent in the intensive care unit), surgical or necropsy findings, and outcome (ie, discharged from hospital or died or was euthanized because of severity of signs or perceived prognosis) was recorded on a spreadsheet. Venous blood gas and plasma lactate concentration variables were measured by use of an amperometric autoanalyzer.a Dogs were presumed to have gastric necrosis if a gastric resection was performed at surgery or if necrosis was identified at the time of necropsy. Survivors were defined as dogs that underwent surgery and were discharged alive from the hospital; nonsurvivors were defined as dogs that underwent surgery and subsequently died or were euthanized or dogs that died or were euthanized before surgery but underwent necropsy.

Statistical analysis—Continuous variable data were assessed for normality via visual inspection of scatterplots and the Shapiro-Wilk test. Depending on whether the data were or were not normally distributed, continuous variables were reported as mean ± SD or median (range), respectively Similarly, the unpaired t test or Wilcoxon rank sum test was used to make comparisons of continuous variables between groups when the data distributions were or were not normal, respectively Proportions and percentages were used to describe categorical variables. The Spearman rank correlation was used to assess the association between base excess and plasma lactate concentration. Receiver operating characteristic curves were generated to assess the diagnostic (for gastric necrosis) and predictive (for outcome) properties of plasma lactate concentration and base excess. The areas under the ROC curves were calculated via the trapezoid method. Exact 95% CIs were calculated for sensitivity and specificity of plasma lactate concentration for predicting gastric necrosis or outcome. Values of P < 0.05 were considered significant for all comparisons. All statistical analyses were performed with a commercially available statistical software package.b

Results

Medical records were available for 122 dogs treated for GDV during the study period. Seventy-eight dogs met the criteria for inclusion in the study. Of the dogs that were excluded, 22 had incomplete medical records, 11 received treatment with fluids or gastric decompression at another facility prior to initial evaluation, 10 were euthanized prior to surgery and did not have a necropsy performed, and 1 had a confounding medical condition (diabetic ketoacidosis).

Signalment—For the 78 dogs, median age was 8.0 years (range, 1.2 to 16.2 years). Fifty (64%) dogs were male, and 28 (36%) were female; 35 (70%) males and 26 (93%) females were neutered. The most common breeds were German Shepherd Dog (n = 14), Great Dane (9), Labrador Retriever (9), Saint Bernard (8), and Standard Poodle (5). Twenty other breeds were represented by < 5 dogs each.

Surgical or necropsy findings and outcome—Gastric necrosis was identified in 16 (21%) dogs during surgery (n = 12) or necropsy (4). Of the 78 dogs, 65 (83%) survived to hospital discharge, and 13 (17%) died or were euthanized. Of the dogs that died or were euthanized, 5 were euthanized before surgery, presumably because of perceived prognosis, financial concerns, or other unknown reasons, and underwent necropsy. Necropsy revealed that 4 of those 5 dogs had gastric necrosis. Three dogs were euthanized during surgery because of severe gastric necrosis (> 70% of tissue affected). Four dogs were euthanized after surgery, of which 2 had disseminated intravascular coagulation and 1 had acute respiratory distress syndrome; for 1 dog, the reasons for euthanasia were not stated in the medical record, but gastric necrosis (> 50% of tissue affected) was noted in the surgical report. One dog died after surgery from respiratory arrest; no necropsy was performed.

Of the 65 survivors that underwent surgery, 8 had gastric necrosis. Of the 8 nonsurvivors that underwent surgery, 4 had gastric necrosis. The proportion of dogs with gastric necrosis in these 2 groups was significantly (P = 0.021) different. Dogs with gastric necrosis that underwent surgery were 7.1 (95% CI, 1.1 to 45.1) times as likely to not survive as were dogs without gastric necrosis that underwent surgery.

Among dogs that survived to discharge from the hospital, the duration of hospitalization for dogs with gastric necrosis was significantly (P < 0.001) longer (median, 4 days; range, 3 to 8 days) than that for dogs without gastric necrosis (median, 2 days; range, 1 to 7 days). Dogs with gastric necrosis also remained in the intensive care unit for a significantly (P = 0.001) longer period (median, 2.5 days; range, 1 to 7 days) than did dogs without gastric necrosis (median, 1 days; range, 0 to 5 days).

Plasma lactate concentration and relationship of that variable with duration of hospitalization, gastric necrosis, and outcome—Median plasma lactate concentration at the time of hospital admission for the 65 dogs that survived to hospital discharge (4.5 mmol/L; range, 0.8 to 14.4 mmol/L) was significantly (P < 0.001) less than the value for the 8 dogs that underwent surgery but did not survive (7.9 mmol/L; range, 5.6 to 15 mmol/L). Plasma lactate concentration was not correlated with overall duration of hospitalization (P = 0.160) but was correlated with time spent in the intensive care unit (P = 0.002). Median plasma lactate concentration at the time of hospital admission for the 12 dogs with gastric necrosis that underwent surgery (6.95 mmol/L; range, 1.9 to 14.4 mmol/L) was significantly (P = 0.008) higher than the value for 61 dogs without gastric necrosis that underwent surgery (4.5 mmol/L; range, 0.8 to 15 mmol/L; Figure 1).

Figure 1—
Figure 1—

Box-and-whisker plot of plasma lactate concentration (mmol/L) at the time of admission to the hospital in dogs with GDV in which gastric necrosis was (n = 12) or was not (61) detected during surgery. For each box, the horizontal line represents the median value, and the upper and lower boundaries represent the 75th and 25th percentiles, respectively. Whiskers represent the minimum and maximum values, and black circles represent outlier values. *Median value for this group is significantly (P = 0.008) different from that for the other group.

Citation: Journal of the American Veterinary Medical Association 242, 1; 10.2460/javma.242.1.54

Via ROC curve analysis, a cutoff plasma lactate concentration of 7.4 mmol/L was 82% accurate for predicting gastric necrosis and 88% accurate for predicting outcome. For prediction of gastric necrosis, sensitivity was 50% and specificity was 88%; for prediction of outcome, sensitivity was 75% and specificity was 89% (Figure 2).

Figure 2—
Figure 2—

Receiver operating characteristic curve for plasma lactate concentration at the time of admission to the hospital and gastric necrosis in dogs with GDV (n = 78; A) and for plasma lactate concentration at the time of admission to the hospital and outcome in dogs with GDV that underwent surgery (73; B). The optimal plasma lactate concentration for predicting gastric necrosis was 7.4 mmol/L, with a specificity of 88% (95% CI, 78% to 95%) and a sensitivity of 50% (95% CI, 25% to 75%). The overall accuracy was 82%. The optimal plasma lactate concentration for predicting outcome was 7.4 mmol/L, with a sensitivity of 75% (95% CI, 35% to 97%) and specificity of 89% (95% CI, 79% to 95%). The overall accuracy was 88%.

Citation: Journal of the American Veterinary Medical Association 242, 1; 10.2460/javma.242.1.54

Base excess and relationship of that variable with plasma lactate concentration, gastric necrosis, and outcome—Data from the 78 dogs indicated that at the time of hospital admission, plasma lactate concentration and base excess were significantly (P < 0.001) and inversely correlated (Figure 3). Among the 73 dogs that underwent surgical intervention, median initial base excess was significantly (P = 0.011) less in the 8 dogs that did not survive to hospital discharge (median, −6.1; range, −14.9 to −2.9), compared with the value in 65 dogs that did survive (median, −3.4; range, −12.2 to 7). Similarly, initial base excess was significantly (P = 0.032) less in the 16 dogs with gastric necrosis (median, −5.7; range, −12.2 to 0.6), compared with the dogs without gastric necrosis (median, −3.4; range, −14.9 to 7; Figure 4). However, base excess was a poor discriminator for predicting gastric necrosis and outcome (area under the ROC curve, 0.571 and 0.565, respectively; Figure 5). The relationships between initial base excess and duration of hospitalization or time spent in the intensive care unit were not assessed.

Figure 3—
Figure 3—

Scatterplot of values for base excess in relation to plasma lactate concentration at the time of admission to the hospital in 78 dogs with GDV. The correlation between base excess and lactate concentration was significant (P < 0.001).

Citation: Journal of the American Veterinary Medical Association 242, 1; 10.2460/javma.242.1.54

Figure 4—
Figure 4—

Box-and-whisker plot of base excess at the time of admission to the hospital in dogs with GDV in which gastric necrosis was (n = 12) or was not (61) detected during surgery. See Figure 1 for key.

Citation: Journal of the American Veterinary Medical Association 242, 1; 10.2460/javma.242.1.54

Figure 5—
Figure 5—

Receiver operating characteristic curve for reciprocal base excess at the time of admission to the hospital and gastric necrosis in dogs with GDV (n = 78; A) and for reciprocal base excess at the time of admission to the hospital and outcome in dogs with GDV that underwent surgery (73; B).

Citation: Journal of the American Veterinary Medical Association 242, 1; 10.2460/javma.242.1.54

Discussion

For dogs with GDV, gastric necrosis appears to have a strong association with outcome and duration of hospitalization, as confirmed by findings of this and other studies.1–5,7,18 With regard to plasma lactate concentration at the time of admission to hospital in the present study, the cutoff for optimal prediction of outcome or gastric necrosis was 7.4 mmol/L; that value is similar to the finding of another study1 involving 102 dogs with GDV performed at our institution, in which the cutoff initial plasma lactate concentration was 6.0 mmol/L with a reported specificity and sensitivity for prediction of gastric necrosis of 88% and 61%, respectively. For the dogs in the present study, plasma lactate concentration was not associated with overall duration of hospitalization, but was associated with time spent in the intensive care unit. These findings could be a reflection of euthanasia of sick dogs earlier in the course of hospitalization. Resolution of hyperlactatemia, which seems to be another important prognostic indicator in dogs with GDV,8 was not investigated in this retrospective study because of low numbers of repeated measurements and variations in treatments and intervals between measurements.

It has been reported11 that there is a 1-to-1 correlation between serum lactate concentration and base excess, which prompted us to investigate this relationship with the hope that base excess could be used as a surrogate for plasma lactate concentration and have similar predictive properties with regard to gastric necrosis and outcome in dogs with GDV. Unfortunately, this was not borne out in the findings of the present study. Although base excess was associated with gastric necrosis and outcome, the association was not sufficiently consistent for base excess to be used as a predictive marker. In fact, the area under the ROC curve for base excess as a predictor of gastric necrosis and outcome in dogs with GDV was 0.571 and 0.565, respectively. The poor predictive effectiveness is likely a result of the many factors that influence values of base excess, including serum concentrations of sodium, chloride, phosphate, proteins, and unmeasured anions. Also, because base excess is a calculated value determined by blood pH, Pco2, and Hct, the patient's acid-base status, duration of clinical signs, and other physical derangements may affect its value. On the other hand, plasma lactate concentration increases as a result of tissue hypoperfusion, including gastric mucosa ischemia, making it a more reliable predictor of disease severity and tissue death. On the basis of the findings of the present study, the authors cannot recommend that base excess assessment be used to reliably predict gastric necrosis or outcome in dogs with GDV.

The present study had several limitations, the most important of which was the retrospective nature of the data collection. Consistency in blood sample collection at the time of admission to the hospital removed some bias, but types of treatment and degree of resuscitation may have distorted the outcome and clouded the findings. For example, some clinicians may have chosen to treat dogs that had high plasma lactate concentrations more aggressively, potentially improving outcomes; other clinicians, aware of the results of a previous study,1 may have more strongly influenced owners' decisions to euthanize dogs that had high plasma lactate concentrations. This would most likely bias the findings toward the null, diminishing the association between base excess and gastric necrosis or outcome. Additionally, collection of a blood sample from a peripheral venous access point may result in less accurate measurements of plasma lactate concentration and base excess than that achieved from a central access point owing to differences in peripheral tissue perfusion; this may have led to inaccurate measurement of plasma lactate concentrations and distorted the true correlation between plasma lactate concentration and base excess. It is also important to mention that gastric necrosis was determined on the basis of surgical or necropsy findings, and histopathologic confirmation of necrosis was not obtained. A prospective study with data collected specifically to evaluate whether base excess can be used to assess gastric necrosis and outcome in dogs with GDV is warranted; a prospective evaluation of plasma lactate concentration over time in affected dogs may also be indicated.

ABBREVIATIONS

CI

Confidence interval

GDV

Gastric dilatation-volvulus

ROC

Receiver operating characteristic

a.

Stat Profile Critical Care Xpress, Nova Biomedical, Waltham, Mass.

b.

Stata, version 11.0 for Windows, Stata Corp, College Station, Tex.

References

  • 1. de Papp E, Drobatz KJ, Hughes D. Plasma lactate concentration as a predictor of gastric necrosis and survival among dogs with gastric dilatation-volvulus: 102 cases (1995–1998). J Am Vet Med Assoc 1999; 215:4952.

    • Search Google Scholar
    • Export Citation
  • 2. Brourman JD, Schertel ER, Allen DA et al. Factors associated with perioperative mortality in dogs with surgically managed gastric dilatation-volvulus: 137 cases (1988–1993). J Am Vet Med Assoc 1996; 208:18551858.

    • Search Google Scholar
    • Export Citation
  • 3. Brockman DJ, Washabau RJ, Drobatz KJ. Canine gastric dilatation/volvulus syndrome in a veterinary critical care unit: 295 cases (1986–1992). J Am Vet Med Assoc 1995; 207:460464.

    • Search Google Scholar
    • Export Citation
  • 4. Glickman LT, Glickman NW, Pérez CM et al. Analysis of risk factors for gastric dilatation and dilatation-volvulus in dogs. J Am Vet Med Assoc 1994; 204:14651471.

    • Search Google Scholar
    • Export Citation
  • 5. Glickman LT, Lantz GC, Schellenberg DB et al. A prospective study of survival and recurrence following the acute gastric dilatation-volvulus syndrome in 136 dogs. J Am Anim Hosp Assoc 1998; 34:253259.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Muir WW. Acid-base and electrolyte disturbances in dogs with gastric dilatation-volvulus. J Am Vet Med Assoc 1982; 181:229231.

  • 7. Matthiesen DT. The gastric dilatation-volvulus complex: medical and surgical considerations. J Am Anim Hosp Assoc 1983; 19:925932.

    • Search Google Scholar
    • Export Citation
  • 8. Zacher LA, Berg J, Shaw SP et al. Association between outcome and changes in plasma lactate concentration during presurgical treatment in dogs with gastric dilatation-volvulus: 64 cases (2002–2008). J Am Vet Med Assoc 2010; 236:892897.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. DiBartola SP. Introduction to acid-base disorders. In: DiBartola SP, ed. Fluid, electrolyte and acid-base disorders in small animal practice. 3rd ed. St Louis: Saunders Elsevier, 2006;229250.

    • Search Google Scholar
    • Export Citation
  • 10. Weiskopf RB, Fairley HB. Anesthesia for major trauma. Surg Clin North Am 1982; 62:3145.

  • 11. Kovacic JP. Acid-base disturbances. In: Silverstein DC, Hopper K, eds. Small animal critical care medicine. St Louis: Saunders Elsevier, 2009;249254.

    • Search Google Scholar
    • Export Citation
  • 12. Davis JW. The relationship of base deficit to lactate in porcine hemorrhagic shock and resuscitation. J Trauma 1994; 36:168172.

  • 13. Davis JW, Shackford SR, Holbrook TL. Base deficit as a sensitive indicator of compensated shock and tissue oxygen utilization. Surg Gynecol Obstet 1991; 173:473476.

    • Search Google Scholar
    • Export Citation
  • 14. Dunham CM, Siegel JH, Weireter L et al. Oxygen debt and metabolic acidemia as quantitative predictors of mortality and the severity of the ischemic insult in hemorrhagic shock. Crit Care Med 1991; 19:231243.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15. Siegel JH, Rivkind AI, Dalal S et al. Early physiologic predictors of injury severity and death in blunt multiple trauma. Arch Surg 1990; 125:498508.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16. Husain FA, Martin MJ, Mullenix PS et al. Serum lactate and base deficit as predictors of mortality and morbidity. Am J Surg 2003; 185:485491.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17. Mikulaschek A, Henry SM, Donovan R et al. Serum lactate is not predicted by anion gap or base excess after trauma resuscitation. J Trauma 1996; 40:218222.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Beck JJ, Staatz AJ, Pelsue DH et al. Risk factors associated with short-term outcome and development of perioperative complications in dogs undergoing surgery because of gastric dilatation-volvulus: 166 cases (1992–2003). J Am Vet Med Assoc 2006; 229:19341939.

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 546 0 0
Full Text Views 1414 978 125
PDF Downloads 902 389 43
Advertisement