|Year : 2014 | Volume
| Issue : 5 | Page : 538-542
Effect of various drugs used in conservative therapy of hepatorenal syndrome: A retrospective drug utilization study
Sharanabasayyaswamy B Hiremath, Mohammed Ameeruddin Kamdod
Department of Pharmacology, Sri Dharmasthala Manjunatheshwara College of Medical Sciences and Hospital, Dharwad, Karnataka, India
|Date of Submission||12-Apr-2013|
|Date of Decision||02-Sep-2013|
|Date of Acceptance||20-Aug-2014|
|Date of Web Publication||11-Sep-2014|
Sharanabasayyaswamy B Hiremath
Department of Pharmacology, Sri Dharmasthala Manjunatheshwara College of Medical Sciences and Hospital, Dharwad, Karnataka
Source of Support: None, Conflict of Interest: None
Objectives: To analyze the impact of various drugs used in conservative therapy on renal failure and mortality in hepatorenal syndrome (HRS) at a tertiary care teaching hospital.
Materials and Methods: Retrospective review of hospital admission records was conducted for case records with HRS as diagnosis. Demographic and clinical data and drug utilization pattern were collected in a pre-designed patient information sheet. Impact of various drugs especially hepatoprotector antioxidant, silymarin, on survival benefits in terms of number of patients alive, change in mean arterial pressure (MAP) and change in serum creatinine at the end of treatment period were estimated by univariate and followed by multivariate analysis.
Results: Of the total 89 case records, 31 met the eligibility criteria and were included in the analysis. On multivariate analysis, a significant correlation between use of intravenous fluids (IVFs) and survival benefits was observed (P < 0.05); wherein patients treated with IVFs had an increase in log odds of survival by 2.42 (95% CI = 1.06 to 121.13) as compared to patient not treated with IVF. However, MAP was not affected by any of the treatment modalities. While change in serum creatinine level was not significantly (P = 0.06, regression correlation = −0.3) correlated with duration of treatment with IVFs.
Conclusion: Use of IVFs may be associated with better short-term survival benefits and favor HRS reversal. Use of silymarin as hepatoprotector antioxidant has no beneficial effects on HRS reversal or survival benefits.
Keywords: Conservative therapy, Drug utilization study, hepatorenal syndrome
|How to cite this article:|
Hiremath SB, Kamdod MA. Effect of various drugs used in conservative therapy of hepatorenal syndrome: A retrospective drug utilization study
. Indian J Pharmacol 2014;46:538-42
|How to cite this URL:|
Hiremath SB, Kamdod MA. Effect of various drugs used in conservative therapy of hepatorenal syndrome: A retrospective drug utilization study
. Indian J Pharmacol [serial online] 2014 [cited 2020 Jan 27];46:538-42. Available from: http://www.ijp-online.com/text.asp?2014/46/5/538/140590
| » Introduction|| |
Hepatorenal syndrome (HRS) occurs in patients with cirrhosis, ascites and hepatic failure. It is characterized by impaired renal function and marked abnormalities in the arterial circulation and activity of endogenous vasoactive system.  Based on clinical features and prognosis, HRS is classified into two types. Type-1 HRS is the most common and severe variety is characterized by acute onset and rapid progression with high mortality rate ranging from 70-90% and mean survival period of two weeks after the onset of renal failure. On the other hand, type-2 HRS has slow progression and comparatively better prognosis. , Among the pharmacological treatments available, combination of systemic vasoconstrictor terlipressin and plasma expander albumin is perhaps the most efficacious in reversing renal failure in HRS. ,
However in Indian hospitals, the use of terlipressin and albumin in HRS at their usual dose regimen may be restricted by poor economic status of the patients.  Under such circumstances, conservative therapy remains as the management approach for HRS. There are no studies analyzing the beneficial or harmful effect of these drugs in conservative therapy. In addition, reversal benefits of N-acetyl cysteine (NAC), a hepatoprotector antioxidant in HRS, support possible benefits associated with conservative therapy.  With this background, the present retrospective study was conducted with the objective of analyzing the drug utilization pattern and impact of various drugs used in conservative therapy and benefits of another hepatoprotector antioxidant, silymarin, in HRS.
| » Materials and Methods|| |
The study was conducted on patients admitted to the tertiary care teaching hospitals of north Karnataka, India, from 1 st January 2007 to 31 st December 2011. After the approval from Institutional Ethics Committee, a retrospective review of hospital admission records was conducted in central medical records department for case records with ICD-10 code K76.7 or HRS as final diagnosis. Case records with HRS as final diagnosis were analyzed for their eligibility to be included under the study based on following predefined eligibility criteria. Inclusion criteria for eligibility included patients of either gender aged more than 18 years with presence of five international ascites club (IAC) framed diagnostic criteria and duration of treatment of more than two days. However, cases with cirrhosis of liver, incomplete case records and patients treated with vasoconstrictors like terlipressin, somatostatin or its analogs, noradrenaline or pressor doses of dopamine for more than 2 days were excluded. As majority of the patients were treated with silymarin as hepatoprotector antioxidant, with the aim of avoiding bias, patients receiving antioxidant other than silymarin were excluded. Patients with baseline serum creatinine >2.5 mg/dl or doubling of baseline serum creatinine to greater than 2.5 mg/dl in less than 2 weeks from the day of admission were considered as type-1 HRS.
Effects of various drugs on survival benefits in terms of number of patients' alive, HRS reversal efficacy in terms of change in serum creatinine level and hemodynamic effects as change in mean arterial pressure (MAP) level at the end of the treatment period were the three outcome measures analyzed. In addition, we also analyzed influence of various demographic and clinical factors on these three outcome measures.
Data were collected by the first author in a pre-designed patient information sheet approved by Institutional Ethics Committee. Baseline demographic and clinical data were collected from the case records. Demographic and clinical factors included as variables were those which have been known to have significant impact on the three outcome measures. , This includes age, type of HRS, alcoholic liver disease (ALD) as cause of cirrhosis, baseline MAP level, baseline serum bilirubin level, baseline serum creatinine level, baseline Child-Pugh score and others. Difference between serum creatinine levels recorded on the day of admission (baseline serum creatinine level) and on the last day of treatment period or the day on which last recorded value was available was considered for estimation of change in serum creatinine level. MAP, calculated as diastolic blood pressure plus one-third of the pulse pressure recorded at the time of admission (baseline MAP) and on the day of discharge from hospital or the day before death, was used for estimation of change in MAP. Final outcome of treatment in terms of death at the end of treatment period was collected according to intention-to-treat (ITT) approach wherein patients discharged with no clinical and or laboratory improvement were assigned the worst possible outcome of being dead. All the drugs and interventions used during the treatment period were collected in terms of dosage form, dose, regimen and duration of treatment. Patients were considered as treated with any drug only when the duration of the treatment with the drug was more than 2 days except for use of intravenous fluids (IVFs), serum albumin infusion and blood transfusion considering significant pathophysiological alterations in fluid homeostasis in patients with cirrhosis.  Few patients who were treated with diuretic torsemide were extrapolated to 'furosemide' considering dose of torsemide to be three times the furosemide.  Considering the significant role of albumin infusion in patients of cirrhosis of liver, we considered the patients receiving blood transfusion as albumin-treated group.  Patients receiving blood transfusion were considered as receiving albumin transfusion and each pint (350 ml) of blood transfused was extrapolated to 15.75 g of albumin infusion considering the mean serum albumin level of a healthy adult blood to be 4.25g
dl (normal range 3.5-5.5g/dl).  As we could not contact all the patients, idea to collect follow-up data for the survival benefits analysis was dropped out to avoid bias.
Baseline demographic and clinical features and drug utilization pattern were described by descriptive statistics. Impact of various drugs, apart from the baseline and clinical factors proved to have a significant effect on the three outcome measures, , which was analyzed by univariate analysis. Pearson's correlation (r) was used for parametric variables with linear distribution and Spearman's correlation (rs) when two variables were nonparametric or parametric with non-linear distribution. This was followed by multiple linear regression analysis (Enter method) for outcome measures change in MAP level and change in serum creatinine level and multiple logistic regression analysis (Enter method) for survival benefit analysis with inclusion of variables, which was found to be significant on univariate analysis (predictor variables) of each outcome measure. Collinearity between the predictor variables of each outcome measures was analyzed by collinearity statistics, Variance Inflation Factor (VIF) value. A value >5 was considered as an indicator of significant co-linearity and under such cases only the predictor variable with strong correlation was included under analysis. Measure of the "Goodness of Fit" of the model for linear regression was estimated by R 2 value along with reporting of ANOVA 'F' statistics and significance (P) value. For logistic regression, it was by Hosmer and Lemeshow test and Nagelkerke pseudo-R 2 value. Sample size required for the study was not determined as there was no predefined 'estimated effect size' value for an intervention or drug and also analysis of impact was not restricted to a single drug or intervention. Statistical analysis was conducted using SPSS version 20.
| » Results|| |
Of the total 89 case records with diagnosis of HRS, 31 met the eligibility criteria. Of the 58 excluded cases, one was excluded for use of antioxidant other than silymarin and five for use of vasoconstrictors (octreotide or somatostatin for upper gastrointestinal tract bleeding) and remaining for not abiding with IAC criteria for diagnosis of HRS. However, in none of the case records there was documentary evidence with regard to abiding to one of the IAC criteria of 'no improvement of serum creatinine (decrease to a level of ≤1.5 mg/dl) after at least two days of diuretic withdrawal and volume expansion with albumin' while diagnosis of HRS. Presuming to be followed, data analysis results of the study are presented here. Of the 31 eligible patients, 11 patients (35.5%) were alive at the end of the mean treatment period of 9.54 ± 4.71 days. Except for one patient, all were male. The ratio of patients with type-1 HRS to type-2 HRS was 27:4. Majority of the patient had ALD (26), followed by chronic HBV (1) and HCV (1) infection. In three patients, the cause was unclear or not recorded.
Baseline clinical and demographic features along with their correlation with three outcome measures are presented in [Table 1]. As there was wide variation in use of antioxidant silymarin, diuretics, IVFs, albumin infusion and blood transfusion, these treatment modalities were included as variables in statistical analysis. [Table 2] shows the mean dose and duration of treatment of these major therapies. Use of multivitamins, laxatives, vitamin-K, proton pump inhibitors and antibiotics (third generation cephalosporins with metronidazole or ornidazole) was universal in all patients with approximately similar dosage regimen and hence were not included as variables and excluded from analysis. Lactulose was used as a laxative at a mean dose of 82.96 ± 30.23 ml/d for mean duration of 5.57 ± 4.34 days. Antibiotics used were a combination of third generation cephalosporins (29 patients) or amoxicillin-clavulanic acid (2 patients) with metronidazole or ornidazole in all the patients. Use of multivitamins as infusions along with IVFs, injection vitamin-K and proton pump inhibitors (pantoprazole or esomeprazole) was universal in all patients with approximately similar dosage regimen.
|Table 1: Baseline demographic and clinical features and their correlation with all three outcome measures|
Click here to view
On univariate analysis, variables found to be significantly correlated with change in MAP level were baseline MAP level (P = 0.005, r = -0.487) and sex (P = 0.042, r = 0.367 for male sex). None of the drugs or treatment modalities had any significant correlation with change in MAP level [Table 2]. With regard to change in serum creatinine level, baseline serum bilirubin level (P = 0.03, r = 0.463), volume of IVFs infused (P = 0.007, r = −0.816) and duration of treatment with IVFs were significantly correlated (P = 0.001, r = −0.888). There was significant correlation between survival and baseline serum bilirubin level (P = 0.018, rs = −0.423) and use of IVFs (P = 0.012, rs = −0.462).
On multiple linear regression, correlation between change in serum creatinine level and its predictor variables was insignificant. There was no significant collinearity between baseline bilirubin level and duration of treatment with IVFs or volume of IVFs used (VIF = 1.32). However collinearity between volume of IVFs and its duration treatment with IVF used was significant (VIF = 6.98). After exclusion of the volume of IVFs as predictor, correlation between change in serum creatinine level (Y) with duration of treatment with IVFs (X 1 ) and baseline bilirubin (X 2 ) was insignificant (regression correlation = −0.3, P = 0.06) (Y = 1.603-0.3 x 1 + 0.031 x 2 ; R 2 = 0.804; F (2,6) =12.32, P = 0.008). After exclusion of the duration of treatment with IVFs as predictor, correlation between change in serum creatinine level (Y) with volume of IVFs used (X 1 ) but not with baseline bilirubin (X 2 ) though significant (Y = 1.566 + 0.0 X 1 + 0.026 X 2 ; R 2 = 0.677; F (2,6) =6.27, P = 0.034) had no impact (regression correlation = 0.000, P = 0.029). The major drawback of the results on change in serum creatinine level is decrease in sample size due to inclusion of only seven patients under analysis. The correlation between change in MAP level (Y) and both of its predictor variables, baseline MAP (X 1 ) and male sex (X 2 ), remained significant (Y = 14.34-0.631 X 1 + 36.24 X 2 ; R 2 = 0.340; F (2, 28) =7.2, P = 0.003) on multivariate analysis. Unlike the previous outcome measure, there was no significant collinearity between these two predictor variables (VIF = 1.01) and alteration in sample size. With regards to survival benefits, on multiple logistic regression correlation between survival was insignificant with baseline bilirubin level and remained significant with use of IVFs (P = 0.044). Use of IVFs was associated with increase in survival benefits by log odds of 2.42 (95% CI = 1.06 to 121.13) than in those not treated with IVFs [Table 3].
|Table 3: Multiple logistic regression analysis model for survival benefits analysis|
Click here to view
| » Discussion|| |
Overall results of the study suggest that the conservative approach to the management of HRS with occasional use of 'albumin' may increase log odds of short-term survival by 2.43 in patients treated with IVFs. With regard to change in serum creatinine level, long duration of treatment and not the volume of IVF may have beneficial effects on HRS reversal. Statistical observations strongly suggest that the use of IVFs is an important predictor of survival benefits in HRS. However, this observation may be controversial in the pathophysiology of volume overload in cirrhosis; hence, the optimum volume of IVFs to be infused remains to be clarified. Considering the significant correlation between change in MAP, change in serum creatinine level and mortality rate, we analyzed the correlation between these outcome measures by linear regression. , Interestingly correlation between change in serum creatinine level (Y) and survival benefits (X) was significant. However, correlation between change in MAP (Y) and survival benefits (X) was insignificant. Similarly correlation between change in MAP (Y) and change in serum creatinine level (X) was also insignificant. With regard to the role of vasoconstrictor terlipressin in HRS, it is found that its survival benefits are significantly influenced by HRS as cause of death and its HRS reversal efficacy by the amount of increase in MAP level achieved at the end of the treatment period. , With this background and in view of the lack of significant effect of IVFs on change in MAP level and its probable beneficial impact on survival benefits and HRS reversal, increased survival benefits associated with use of IVFs can be correlated to their favorable HRS reversal effects in absence of significant effects on hemodynamic parameters. Similar observations were noted in the study on role of NAC in HRS. Though the biological plausibility behind the beneficial effects of NAC in HRS can be explained on the basis of its hepatoprotector action, same is not applicable to IVFs. It can be hypothesized that rather than acute plasma volume expansion, continuous slow infusion of low volume of IVFs may help in maintaining the perfusion of organs especially kidney and thus inhibit progression of HRS and improve survival benefits. As there are no controlled studies justifying or reputing the benefits of use of IVFs in HRS and considering the strong statistically significant correlation between survival and use of IVFs, their role in HRS needs to be further analyzed in randomized trials along with the type of IVF that benefits more.
Lack of beneficial effects of silymarin observed in our study are highly relevant in the background of beneficial effects observed with NAC in HRS. Reason behind the need for and significance of studies on role of antioxidants in HRS is the issue of non-responder state to terlipressin, the major unmet need in the pharmacological treatment of HRS. , Role of oxidative stress in vascular non-reactivity and thus non-responder state to vasoconstrictors in HRS and its association with multi-organ failure, the major non-HRS related cause of death in HRS patients treated with terlipressin signifies the need for studies on role of antioxidants in HRS.  Reported beneficial effects of NAC in HRS are from an uncontrolled study with major drawback of lack of control group.  The present retrospective drug utilization study is thus significant with regard to this drawback as it included untreated patients equivalent to control group. However as per the results of our study benefits of silymarin on change in MAP level, HRS reversal or survival rate in HRS are doubtful at the dose and duration for which it is used in our study. Unlike the evidences from NAC study, which reports significant HRS reversal benefits in absence of its significant effects on hemodynamic parameters, lack of HRS reversal benefits of silymarin could be explained on the hypothesis of insufficient dose and duration of treatment with silymarin. Nevertheless, considering the uncontrolled nature of the NAC study design and results of our study, we are of the opinion that the survival benefits and HRS reversal efficacy of silymarin and perhaps other antioxidants alone are doubtful and needs to be studied further under randomized design.
Major strength of our study is analysis of impact of various drugs used in conservative therapy, especially the effect of another hepatoprotector antioxidant silymarin in HRS. Though analysis is arbitrary, we hope results give a relative idea on the impact of individual drugs on the three major outcome measures of HRS. Patients included in our study belonged predominantly to type 1 HRS, and the analyzed HRS reversal efficacy and survival benefits are not diluted by the inclusion of patients with favorable demographic and clinical features. Our analysis of survival benefits was also stringent with ITT approach. However in addition to the inherent drawbacks of any observational study, our study has few more limitations. The extrapolations made with regard to dose of 'furosemide' used and dose of 'albumin' infused may not be strictly justifiable, but it was necessary to avoid bias. In case of statistical aspects, small sample size (small number of events per variable) may not validate statistical inferences on the observed effects of various drugs used in this study, including silymarin. However considering the low incidence of HRS, sample size of 31 appears to be justifiable. In addition, considering only two co-variants (baseline serum bilirubin level and use of IVFs) having significant correlation with survival benefits, the required number of cases (>10) per co-variant appears sufficient.  The major drawback is the uncertainty over diagnosis of HRS with regard to abiding to IAC diagnostic criteria of 'no improvement of serum creatinine level after at least two days of diuretic withdrawal and volume expansion with albumin.' Whether withdrawal of diuretic and use of albumin at dose of 1 g/kg body weight per day as per revised criteria or intravenous fluids at dose of 1.5 L/day as per previous criteria were followed at the time of diagnosis of HRS is uncertain. The observed strong correlation between improved survival benefits and use of IVFs can also be explained on this drawback. However as stated earlier, we could not found evidences to refute or accept this hypothesis.
To conclude, use of IVFs may be associated with improved short-term survival benefits. Increase in the duration of treatment with IVFs may favor HRS reversal benefits. Considering the limitations of the study, result of our study needs to be confirmed in controlled trials.
| » Acknowledgement|| |
We are thankful to Dr. Niranjan Kumar Medical Director and Dr. J.V. Chowti principal, SDM college of Medical Sciences and Hospital Dharwad for their help and support.
| » References|| |
|1.||Fernandez J, Arroyo V. Novel definition of hepatorenal syndrome: Clinical consequences. Front Gastrointest Res 2011;28:122-9. |
|2.||Wadei HM, Mai ML, Ahsan N, Gonwa TA. Hepatorenal syndrome: Pathophysiology and management. Clin J Am Soc Nephrol 2006;1:1066-79. |
|3.||Gluud LL, Christensen K, Christensen E, Krag A. Systematic review of randomized trials on vasoconstrictor drugs for hepatorenal syndrome. Hepatology 2010;51:576-84. |
|4.||Holt S, Goodier D, Marley R, Patch D, Burroughs A, Fernando B, et al. Improvement in renal function in hepatorenal syndrome with N-acetylcysteine. Lancet 1999;353:294-5. |
|5.||Hiremath SB, Srinivas LD. Survival benefits of terlipressin and non-responder state in hepatorenal syndrome: A meta-analysis. Indian J Pharmacol 2013;45:54-60. |
|6.||Boyer TD, Sanyal AJ, Garcia-Tsao G, Blei A, Carl D, Bexon AS, et al.; Terlipressin Study Group. Predictors of response to terlipressin plus albumin in hepatorenal syndrome (HRS) type 1: Relationship of serum creatinine to hemodynamics. J Hepatol 2011;55:315-21. |
|7.||Jackson EK. Diuretics. In: Laurence LB, John SL, Keith LP, editors. Goodman and Gillman's The Pharmacological basis Therapeutics. 11 th ed. New York: McGraw-Hill; 2008. P. 737-69. |
|8.||Lee JS. Albumin for end-stage liver disease. Korean J Intern Med 2012;27:13-9. |
|9.||Sridevi D. Medscape reference drugs, diseases and procedures: Albumin. 2012 Jan. Available from: http://emedicine.medscape.com/article/2054430-overview [Last accessed on 2013 Feb 1]. |
|10.||Kumar R, Indrayan A, Chhabra P. Reporting quality of multivariable logistic regression in selected Indian medical journals. J Postgrad Med 2012;58:123-6. |
[Table 1], [Table 2], [Table 3]