J Nephrol DOI 10.1007/s40620-014-0152-2
ORIGINAL ARTICLE
Acute kidney injury after cardiac surgery: is minocycline protective? Ladan Golestaneh • Kathryn Lindsey • Pooja Malhotra • Faraj Kargoli • Emily Farkas • Hendrick Barner • Rizwan Qazi • Anna Schmidt • Michael Rauchman • Ziyad Al-Aly • Robert Johnson • Kevin Martin • Pierre Dagher • Allon Friedman • Tarek M. El-Achkar
Received: 6 August 2014 / Accepted: 13 October 2014 Ó Italian Society of Nephrology 2014
Abstract Background and objectives Acute kidney injury (AKI) after cardiac bypass surgery (CABG) is common and carries a significant association with morbidity and mortality. Since minocycline therapy attenuates kidney injury in animal models of AKI, we tested its effects in patients undergoing CABG. Design, setting, participants and measurements This is a randomized, double-blinded, placebo-controlled, multicenter study. We screened high risk patients who were scheduled to undergo CABG in two medical centers between Jan 2008 and June 2011. 40 patients were randomized and 19 patients in each group completed the study. Minocycline prophylaxis was given twice daily, at least for four doses prior to CABG. Primary outcome was defined as AKI [0.3 mg/dl increase in creatinine (Cr)] within 5 days after surgery. Daily serum Cr for 5 days, various clinical and hemodynamic measures and length of stay were recorded. Results The two groups had similar baseline and intraoperative characteristics. The primary outcome occurred in 52.6 % of patients in the minocycline group as compared to L. Golestaneh (&) P. Malhotra F. Kargoli Montefiore Medical Center, Albert Einstein Medical Center, 3411 Wayne Ave, Suite 5H, Bronx, NY 10467, USA e-mail: [email protected] K. Lindsey E. Farkas H. Barner R. Qazi A. Schmidt M. Rauchman R. Johnson K. Martin St Louis University Hospital, St Louis, MO 63110, USA Z. Al-Aly VA St Louis Health Care System, St Louis, MO 63106, USA P. Dagher A. Friedman T. M. El-Achkar IU Health University Hospital, Indianapolis, IN 46202, USA
36.8 % of patients in the placebo group (p = 0.51). Peak Cr was 1.6 ± 0.7 vs. 1.5 ± 0.7 mg/dl (p = 0.45) in minocycline and placebo groups, respectively. Death at 30 days occurred in 0 vs. 10.5 % in the minocycline and placebo groups, respectively (p = 0.48). There were no differences in post-operative length of stay, and cardiovascular events between the two groups. There was a trend towards lower diastolic pulmonary artery pressure [16.8 ± 4.7 vs. 20.7 ± 6.6 mmHg (p = 0.059)] and central venous pressure [11.8 ± 4.3 vs. 14.6 ± 5.6 mmHg (p = 0.13)] in the minocycline group compared to placebo on the first day after surgery. Conclusions Minocycline did not protect against AKI post-CABG. Keywords Acute kidney injury Cardiac surgery Minocycline
Introduction Acute kidney injury (AKI) after cardiac surgery remains a serious challenge [1]. The reported incidence of this complication depends on the definition used for AKI, but can reach as high as 30 %. The development of AKI is associated with increased in-hospital morbidity and mortality, and affects long-term survival [2]. Multiple interventions have attempted to prevent AKI after cardiac surgery but so far with disappointing results. Most of these interventions have centered on mitigating ischemia and inflammation, features of acute tubular necrosis (ATN), the presumed etiology of AKI in this setting. Risk factors for the development of postoperative AKI include patient related factors such as pre-existent kidney dysfunction, female gender, reduced left ventricular
123
J Nephrol
function, peripheral vascular disease, diabetes mellitus; and surgery related factors: including length of surgery, crossclamp time, off-pump versus on-pump, nonpulsatile flow, valvular surgery, hemolysis and hemodilution [4, 7, 9, 10]. The pathogenic basis of ATN was once thought to be a state of renal hypo-perfusion and resultant tissue ischemia [11]. However recently we have come to understand that ATN is downstream of earlier, highly complex events [12]. Injured endothelial cells initiate patterns of insult resulting in microvascular injury, leak and coagulation derangements. These processes lead to the activation of bonemarrow derived cells, various cytokines and interleukins with damage to endothelial and renal epithelial cells. Cardiac bypass is associated with activation of inflammatory cytokines and complement [13–17]. Tetracyclines possess a wide repertoire of anti-inflammatory and immunomodulatory actions in addition to their well-characterized antimicrobial effects [18, 19]. There is growing experimental evidence that tetracyclines such as minocycline, have a beneficial effect on cardiac myocytes and infarct size, the mechanism for which involves inhibition of free radical and cytokine production, interference with proteolysis, and modulation of matrix metalloproteinase activity [18–29]. Tetracyclines also possess potent antiapoptotic properties manifested by inhibition of caspase 1 and 3 expression and direct blockade of cytochrome c release from mitochondria [23, 24]. The potential benefits of minocycline and its high uptake by cardiac myocytes make it an ideal candidate for use in the setting of cardiac bypass surgery. This pilot study aimed at testing the hypothesis that pretreatment with minocycline of patients at high risk for AKI undergoing cardiac bypass surgery protects against developing acute kidney injury and improves overall cardiac outcomes.
Materials and methods Approval for this pilot study (ClinicalTrials.gov registration: NCT 00556491) was obtained from the St. Louis University and Montefiore Medical Center IRB committees. We adhered to the Declaration of Helsinki and informed consent was obtained from all subjects. Study design and patient population This study was designed as a double blinded, randomized, placebo controlled trial recruiting patients undergoing cardiac bypass surgery at Saint Louis University hospital (Missouri) and Montefiore Medical Center (New York). Patients were randomized to receive either minocycline capsules, 200 mg initially, then 100 mg every 12 h until
123
surgery, or placebo. Minocycline and placebo capsules looked identical and were provided by Triax. Both investigators and patients were blinded to the randomization, which was performed using permutated block method. The total number of doses, including the loading dose was at least 4 (i.e. 36 h prior to surgery minimum, last dose early morning on surgery day) to a maximum of 14 doses (7 days prior to surgery maximum). Patients were screened as soon as they were scheduled for bypass surgery. If eligible, and after informed consent was granted, they were randomized to either minocycline or placebo until surgery. Patients otherwise received standard medical and surgical care. Most of the parameters that were collected are part of the standard care of these types of surgical patients. Recruitment for the study started in January 2008 and finished in June 2011. The study population included all patients who were scheduled to undergo cardiac bypass surgery at both centers. They were patients with ischemic heart disease and/or valvular abnormalities who were scheduled to undergo open heart surgery (coronary artery bypass surgery with or without valvular repair/replacement). We did not target a specific age group, ethnicity or gender. However, the location and the disease entity of our study population excluded children a priori. To increase the event rate, we initially targeted patients with chronic kidney disease (CKD 2-4) calculated by MDRD formula. However, because of slow recruitment, and the fact that our urban centers delivered care to patients with a higher burden of chronic diseases such as diabetes, hypertension, and vascular disease (i.e. a high risk group), we removed the GFR limitation after year one of the study to include all ranges of GFR, excluding CKD 5. Our study did not include any subject population deemed as vulnerable. The inclusion and exclusion criteria are presented in Table 1. Verification that the patients took the medication was done by direct observation for inpatients. For patients who were randomized as outpatients, this was done by counting the remainder of the pills the day of surgery. Data collection Baseline demographic data and chronic medical illnesses for each patient enrolled in the study were recorded at randomization. Enrolled patient were monitored closely for signs and symptoms of any adverse effects of minocycline like allergic reactions, headaches, vertigo, dizziness and photosensitivity. In addition, major complications, and medication history were noted in detail. We monitored daily clinical and laboratory data needed for the study as described in the timeline, and labs were ordered in case they were missed by the primary team. We also collected hemodynamic data the 1st day when all patients had Swan-
J Nephrol Table 1 Baseline characteristics Minocycline (N = 19)
Placebo (N = 19)
Age (years)
64.1 ± 9.9
60.8 ± 11.5
p
0.34
White (%)
47.4
52.6
[0.9
African-American (%)
36.8
36.8
[0.9
Hispanic (%)
15.8
10.6
[0.9 [0.9
Male (%)
78.9
73.7
DM (%)
57.9
47.4
COPD (%) PVD (%)
15.8 31.6
15.8 36.8
CKD (%)
36.8
26.3
HTN (%)
94.7
94.7
[0.9
CHF (%)
47.4
42.1
[0.9 [0.9
0.75 [0.9 [0.9 0.73
EF B35 % (%)
26.3
26.3
Baseline Cr (mg/dl)
1.2 ± 0.5
1.1 ± 0.3
0.41
BMI
28.8 ± 3.6
32.4 ± 7.7
0.08
Albuminuria (%)
26.3
15.8
0.69
GFR (ml/min/1.73 m2)
71.9 ± 26.4
77.3 ± 27.4
0.54
Drug dose
6.4 ± 3.1
6.7 ± 3.3
0.78
Preop infection or antibiotics (%) Postop antibiotics (%)
0
10.0
0.15
68.4
47.4
0.90 0.08
Post op NSAIDS (%)
26.3
5.2
ACE I or ARB (%)
68.4
68.4
1.0
Diuretic use (%)
36.8
42.1
0.74
Statin use (%)
68.4
68.4
1.0
Beta blocker (%)
84.2
68.4
0.25
Calcium channel blocker (%)
26.3
42.1
0.30
Ganz catheter, detailed daily medication administration, fluid intake and output, ventilator use, length of stay in the ICU, and total hospital stay after surgery. Since 30 day mortality is a secondary objective in our study, a follow up phone call was made to patients to check their mortality status. Definitions In addition to the definitions discussed above we used the following other preset definitions: Baseline Cr was defined as the most recent Cr which typically was at the time of randomization. Our pre-specified primary outcome was the occurrence of AKI, defined as 0.3 mg/dl increase in Cr within any 48 h time period after surgery and up to a maximum of 5 days post-operatively. This definition is consistent with the acute kidney injury network (AKIN) criteria. Chronic kidney disease (CKD) was defined as an eGFR \60 ml/min as estimated by the 4-variable MDRD study equation. Secondary outcomes that were also reported were death up to 30 days post-op, cardiovascular
events, infections, and length of stay. We also examined cardiac hemodynamic parameters (cardiac index, systemic vascular resistance, pulmonary artery diastolic and central venous pressures) uniformly recorded at 6 a.m. on the first day after surgery from Swan-Ganz catheters. Data analysis Baseline and intra-operative variables of the control and intervention group were tabulated. A Fisher’s exact test was used to compare proportions for categorical variables and a t test was used to examine the difference in means for continuous data (reported as mean ± standard deviation). Univariate and regression analyses were performed as well to derive any significant associations with primary and secondary outcomes. Baseline variables included in this analysis were demographics (age, race), baseline kidney function and presence of albuminuria; chronic illnesses known to be risk factors for AKI after surgery such as diabetes mellitus (DM), hypertension (HTN), peripheral vascular disease (PVD), congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), body mass index (BMI); various medications by class such as diuretics, beta blockers, angiotensin converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARB) and statins. Intra-operative variables that were compared between the two groups included: surgery duration, crossclamp time, bypass (pump) time, lowest systolic pressure, day 1 lowest BP, day 1 cardiac output, day 1 diuretic use and day 1 statin use. Crude primary and secondary outcomes were compared using Fisher’s exact. A p value less than 0.05 were considered statistically significant. A t test was used to compare the difference in means for all hemodynamic measurements. Comparison of length of stay was done using t test after logarithmic transformation, since its distribution tends to be skewed. We also performed similar analysis on the whole population grouped into those patients who developed AKI vs. those who did not (no-AKI). This allowed us to determine whether certain risk factors were unevenly distributed in patients who developed AKI. Safety monitoring Adverse events were monitored until discharge. An independent Data Safety Monitoring Board (DSMB) had access to the unblended data. All pertinent safety data were reviewed from randomization until 1 week post-op and the serious adverse events (including death) up to 30 days post-op. The DSMB also performed quarterly reviews and advised the study investigators on any actions pertaining to the safety. All DSMB communications were reported to the IRB.
123
J Nephrol
Results Study flow, baseline demographics and intra-operative data We screened consecutive patients scheduled for cardiac surgery in two medical centers in St Louis and Bronx. The most common reason for not recruiting patients was the lack of timely notification to ensure there was enough time (at least 36 h) to give the drug prior to CABG. Another major factor was the emergent status of many surgeries. Other exclusion criteria are listed in Fig. 1. 40 patients were randomized to the study. Surgery was canceled in the case of 1 patient, and another patient withdrew himself from the study after randomization. 19 patients in each group underwent surgery and were included in the analysis. Table 1 shows the baseline characteristics of patients in the minocycline compared to the placebo groups. The two groups had similar baseline characteristics. The average age was 64.1 ± 9.9 in the minocycline group vs. 60.8 ± 11.5 in placebo (p = 0.34). Approximately half of the patients were of white ethnicity, predominantly of male gender. Diabetes was prevalent in 57.9 % of minocycline vs. 47.4 % of placebo groups (p = 0.75). Most of the population studied had HTN, and almost a third of the patients in each group had CKD. The prevalence of CHF in both groups was also high: 47.4 and 42.1 % in minocycline and placebo groups, respectively. Patients in the placebo group trended towards having a higher BMI, but this was not statistically significant (p = 0.08). There was no difference in blood pressure lowering medication intake by class (ACEI/ARB, beta blockers, calcium channel blockers, diuretics) or statin use between the two groups (Table 1). On baseline physical exam: 5.2 % of patient in the placebo
group had crackles, 10.5 % had edema. In the minocycline group: none of the patients had crackles and 15.8 % of patients had edema. On post-op day 1: 68.4 % of patients in the placebo group were given antibiotics (cefazolin and vancomycin); none of the patients were documented as having infection but one of the patients was classified as having a systemic inflammatory response syndrome (SIRS) likely from bypass. In the minocycline group: 47.4 % were given antibiotics consisting of cefazolin and vancomycin and none of the patients had documented infection or sepsis. After randomization, patients in the treatment group received an average of 6.4 ± 3.1 doses of minocycline, which was comparable to the number placebo doses given to the control group (6.7 ± 3.3, p = 0.78). The intraoperative data for patients in both study groups are presented in Table 2. The two groups had similar intraoperative characteristics. There was a trend towards a shorter cross-clamp time in the minocycline group compared to placebo (103.3 ± 39.3 vs. 141.1 ± 72.9, respectively; p = 0.06). The average lowest systolic BP was 79.9 ± 16.5 mmHg in the minocycline group vs. 88.7 ± 13.3 mmHg in placebo (p = 0.09). Duration of nonpulsatile flow in the placebo group was 172 ± 82 min, and 138 ± 57 min in the minocycline group. Primary and secondary outcomes Acute kidney injury, as defined by the AKIN criteria, occurred in 10 (58.3 %) patients in the minocycline group compared to 7 (36.8 %) patients in placebo (p = 0.51). There was no difference in the peak serum Cr between the two groups (Table 1). One patient required dialysis in each group. No patients died in the minocycline group compared to two deaths in the placebo group (p = 0.48). There was
Fig. 1 Serum creatinine levels after cardiac surgery; a comparing patient in the treatment vs. placebo groups and b comparing patient who developed AKI vs. those who did not
123
J Nephrol Table 2 Intra-op characteristics Minocycline (N = 19)
Table 3 Outcomes in each group Placebo (N = 19)
p
Minocycline N (%)
Placebo N (%)
p
Procedure duration (min)
355.5 ± 109.3
395.6 ± 127.2
0.32
AKI
10 (52.6)
7 (36.8)
0.51
Cross-clamp time (min)
103.3 ± 39.3
141.1 ± 72.9
0.06
Peak Cr (mg/dl)
1.6 ± 0.7
1.5 ± 0.7
0.45
Pump time (min)
138.0 ± 57.4
172.6 ± 85.1
0.15
Dialysis
1 (5.2)
1 (5.2)
1.0
79.9 ± 16.5
88.7 ± 13.3
0.09
Death (%)
0 (0)
2 (10.5)
0.48
Post-op hospital days
8.7 ± 3.8
9.8 ± 7.2
0.9
Post-op ICU days
4.8 ± 3.6
6.1 ± 7.1
0.75 0.45
Lowest systolic BP (mmHg) Urine output (ml)
991.5 ± 737.5
1,263.8 ± 888.3
0.34
no difference between the two groups in post-operative hospital and ICU stay (Table 2).
Ventilated [48 h
3 (15.8)
5 (26.3)
Infections
2 (10.5)
3 (15.8)
0.66
Stroke
2 (10.5)
0 (0)
0.48
Re-op
2 (10.5)
1 (5.6)
Minocycline (N = 19)
Placebo (N = 19)
p
92.9 ± 9.8
93.2 ± 12.6
0.94
Minocycline and hemodynamic parameters on day 1 post-operatively We examined whether minocycline treatment had an effect on hemodynamic parameters and cardiac pressures on day 1 after surgery. Table 2 shows that patients in the minocycline and placebo groups had comparable cardiac indices, systemic vascular resistances, and lowest systolic pressures. Patients in both groups received an equivalent amount of fluid, and had a comparable urine output. However, compared to placebo, patients in the minocycline group had a trend towards lower cardiac filling pressures as measured by the pulmonary artery diastolic (PAD) pressure (a surrogate for left ventricular end diastolic pressure): 16.8 ± 4.7 vs. 20.7 ± 6.6 mmHg in minocycline vs. placebo, respectively (p = 0.059). The central venous pressure (CVP) also trended lower in the minocycline group 11.8 ± 4.3 vs. 14.6 ± 5.9 mmHg in placebo (p = 0.13). There was no difference in these two pressures when we performed an analysis factoring BMI: BMI C30 vs. \30 [18.8 vs. 18.7 mmHg, respectively for PAD (p = 0.96); 13.3 vs. 13.1 mmHg, respectively for CVP (p = 0.92)]; thus uneven distribution of BMI between the treatment groups is unlikely to explain this finding. Characteristics of patients with AKI To better understand the characteristics and outcomes of patients who develop AKI, we grouped the patients in our study population based on whether they developed AKI, and examined all the variables (risk factors, intra-op events, and outcomes). Table 3 shows only the significant variables that were unevenly distributed with each group. Patients with AKI were more likely to be obese, and have peripheral vascular disease. They also tend to have longer surgeries, and develop more intra-operative hypotension. Patients with AKI stayed longer in the ICU and in the hospital after surgeries. Figure 1a shows the average serum Cr between the two groups. Figure 1b shows that AKI is most commonly
Lowest systolic BP (mmHg)
[0.9
Fluid Intake (ml)
2,923.6 ± 1,197.5
3,134.5 ± 946.4
0.57
Urine Output (ml)
2,031.4 ± 1,179.6
2,355.8 ± 931.0
0.40
Cardiac Index
2.8 ± 0.7
3.0 ± 0.9
0.53
SVR
1,051.6 ± 342.3
989.6 ± 355.1
0.61
PAD
16.8 ± 4.7
20.7 ± 6.6
0.059
CVP
11.8 ± 4.3
14.6 ± 5.9
0.13
Vasopressor use (%)
58.8
50.0
0.74
Ventilator (%)
52.9
66.7
0.73
BUN
19.2 ± 7.4
16.7 ± 3.9
0.21
Cr
1.3 ± 0.5
1.1 ± 0.3
0.22
detected at day 2 post-operatively and the serum Cr on average peaks at day 3. Using regression analysis the following baseline parameters were significantly associated with developing AKI post-op: baseline PVD, higher BMI, pre-op CKD as marked by abnormal creatinine, and pre-op age. Minocycline safety Minocycline was very well tolerated by all patients. Only one serious adverse event occurred prior to surgery in a patient on chronic anticoagulation, where treatment with minocycline may have contributed to him being supratherapeutic on Coumadin, requiring admission for anemia and gastro-intestinal bleeding, that was treated conservatively. There was no difference in serious adverse events post-operatively, as shown in Table 3.
Discussion In this pilot study, we examined the effect of minocycline prophylaxis on developing AKI after cardiac surgery with
123
J Nephrol
cardio-pulmonary bypass. Our study was a prospective study with comprehensive data gathering including intraop and post-op parameters that are seldom gathered in other studies. We examined secondary outcomes, including risk factors for AKI, using the specific data points derived. Although our pilot study did not find an effect of minocycline on incidence of AKI after CABG, the findings could help design a large and adequately powered trial. The therapeutic effects of tetracyclines on cardiac muscle post-ischemia are time sensitive: it is important to block proteases that contribute to pathologic cell remodeling and not block those that help surrounding tissue recover [20–28]. For example: post-ischemia minocycline treatment was protective in a rodent model of atherosclerotic cardiac ischemia, as was administration of minocycline prior to 36 h before the renal clamping in an AKI murine model [27, 28]. The timing of minocycline administration is an open question with regards to future studies. In any clinical study of AKI, there is debate about the applicability of animal models. The minocycline data is no different. While administration of minocycline pre and/ or post ischemia was shown to be protective in rodent clamp models of ischemic ATN, whether this agent protects against cardiac surgery specific renal injury is unclear [27, 28]. In our study minocycline treatment appeared to be associated with lower cardiac filling pressures despite similar fluid balance and cardiac indices. This suggests that minocycline induces a state of improved cardiac compliance, which could be beneficial for cardiovascular and pulmonary outcomes. The effect of minocycline on cardiac compliance does have a biological explanation. Studies in rodents showed that minocycline rapidly accumulates in the myocardium after administration [18, 27, 28]. Minocycline also had protective effects on the myocardium postinfarction, and inhibited oxidative stress [28]. Tao et al. [29] showed more recently that minocycline protects cardiac myocytes possibly through inhibition of poly(ADPribose) polymerase-1 (PARP-1). Extensive PARP-1 activation can lead to cell death through bio-energetic failure and depletion of ATP. Pharmacologic and genetic inhibition of PARP-1 improved cardiac contractile function in an experimental model of heart failure [29]. Therefore, we propose that minocycline improves cardiac compliance post CABG by its protective effects on the myocardium and possibly though preserving energy stores by PARP-1 inhibition. This finding underscores the importance of performing a larger trial to examine the effect of minocycline prophylaxis on cardiovascular, pulmonary and length of stay outcomes. Significant factors associated with developing AKI in our population were the presence of peripheral vascular
123
disease, age and presence of CKD at baseline. This is consistent with previously published data [2–11]. Moreover, longer surgeries and intraoperative hypotension were also more common in patients with AKI. Examining intraoperative data is essential to determining the risks of AKI in patients undergoing cardiac bypass. The association of AKI with length of stay is in accordance with previous observational data from Chertow et al. [2]. The incidence of AKI in our study population (45.6 %) was higher than reported by previous studies. This could be explained by the use of a very sensitive definition (AKIN stage 1) and the recruitment of a high risk population, where diabetes, HTN, PVD and CKD were highly prevalent. All surgeries by design were performed using bypass, which carries a higher risk for AKI than off pump cardiac surgery. Our study population included a significant proportion of African-American and Hispanic participants. This could also influence the incidence of AKI, but at the same time improve the validity of the findings. With the advent of new diagnostic biomarkers the possibility of detecting AKI earlier and more accurately than what is offered by serum creatinine, may be realized. NGAL has been shown to be of value in the early diagnosis of AKI in the post cardiac surgery setting; however consensus guidelines use serum creatinine for AKI definition. Future projects will incorporate a diverse panel of biomarkers, once they are validated. A major strength of our study is in its design as a randomized, double blinded, placebo controlled study. Our study is unique in providing detailed intraoperative variables, and post-operative data such as hemodynamic measurement. Furthermore, our study recruited patients from two different centers located in two states, and recruited a diverse population, which improved its validity. The major limitation of this study is its sample size, which precludes major conclusions on the primary outcome. However, our study can be used as a model to design future studies to administer prophylactic interventions in patients undergoing cardiac surgery. In conclusion, in our pilot study, minocycline prophylaxis prior to cardiac surgery with bypass did not reduce the occurrence of AKI, but could have a beneficial effect on cardiac compliance. These findings provide a rationale and a roadmap to perform a large trial that investigates not only renal, but also cardiovascular, pulmonary and length of stay endpoints. Conflict of interest This study was supported by a Norman Coplon Grant from Satellite Healthcare, and by a limited grant from Triax Pharmaceuticals (makers of generic form of minocycline).
J Nephrol
References 1. Mehta RL (2005) Acute renal failure and cardiac surgery: marching in place or moving ahead? J Am Soc Nephrol 16(1):12–14 2. Chertow GM et al (1998) Independent association between acute renal failure and mortality following cardiac surgery. Am J Med 104(4):343–348 3. Fortescue EB, Bates DW, Chertow GM (2000) Predicting acute renal failure after coronary bypass surgery: cross-validation of two risk-stratification algorithms. Kidney Int 57(6):2594–2602 4. Slogoff S et al (1990) Role of perfusion pressure and flow in major organ dysfunction after cardiopulmonary bypass. Ann Thorac Surg 50(6):911–918 5. Thakar CV et al (2005) A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol 16(1):162–168 6. Thakar CV et al (2003) ARF after open-heart surgery: influence of gender and race. Am J Kidney Dis 41(4):742–751 7. Tuttle KR et al (2003) Predictors of ARF after cardiac surgical procedures. Am J Kidney Dis 41(1):76–83 8. Chertow GM et al (1997) Preoperative renal risk stratification. Circulation 95(4):878–884 9. Abramov D et al (2003) The influence of cardiopulmonary bypass flow characteristics on the clinical outcome of 1820 coronary bypass patients. Can J Cardiol 19(3):237–243 10. Fischer UM et al (2002) Impact of cardiopulmonary bypass management on postcardiac surgery renal function. Perfusion 17(6):401–406 11. Rosner MH, Okusa MD (2006) Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol 1:19–32 12. Devarajan P (2006) Update on mechanisms of ischemic acute kidney injury. J Am Soc Nephrol 17(6):1503–1520 13. Molitoris BA (2003) Transitioning to therapy in ischemic acute renal failure. J Am Soc Nephrol 14(1):265–267 14. Molitoris BA, Sutton TA (2004) Endothelial injury and dysfunction: role in the extension phase of acute renal failure. Kidney Int 66(2):496–499 15. Okusa MD (2002) The inflammatory cascade in acute ischemic renal failure. Nephron 90(2):133–138 16. Sheridan AM, Bonventre JV (2000) Cell biology and molecular mechanisms of injury in ischemic acute renal failure. Curr Opin Nephrol Hypertens 9(4):427–434
17. Sutton TA, Fisher CJ, Molitoris BA (2002) Microvascular endothelial injury and dysfunction during ischemic acute renal failure. Kidney Int 62(5):1539–1549 18. Griffin MO et al (2010) Tetracyclines: a pleitropic family of compounds with promising therapeutic properties. Review of the literature. Am J Physiol Cell Physiol 299:C539–C548 19. Nieman GF, Zerler BR (2001) A role for the anti-inflammatory properties of tetracyclines in the prevention of acute lung injury. Curr Med Chem 8(3):317–325 20. Arvin KL et al (2002) Minocycline markedly protects the neonatal brain against hypoxic-ischemic injury. Ann Neurol 52(1):54–61 21. Cheung PY et al (2000) Matrix metalloproteinase-2 contributes to ischemia-reperfusion injury in the heart. Circulation 101(15):1833–1839 22. Sloan R et al (2012) Mitochondrial permeability transition in the diabetic heart: contributions of thiol redox state and mitochondrial calcium to augmented reperfusion injury. J Mol Cell Card 52:1009–1018 23. Tikka T et al (2001) Tetracycline derivatives and ceftriaxone, a cephalosporin antibiotic, protect neurons against apoptosis induced by ionizing radiation. J Neurochem 78(6):1409–1414 24. Power C et al (2003) Intracerebral hemorrhage induces macrophage activation and matrix metalloproteinases. Ann Neurol 53(6):731–742 25. Chen M et al (2000) Minocycline inhibits caspase-1 and caspase3 expression and delays mortality in a transgenic mouse model of Huntington disease. Nat Med 6(7):797–801 26. Zhu S et al (2002) Minocycline inhibits cytochrome c release and delays progression of amyotrophic lateral sclerosis in mice. Nature 417(6884):74–78 27. Xia D et al (2011) Adminsitration of minocycline ameliorates damage in a renal ischemia/reperfusion injury model. Clin Invest Med 34(2):E55–E63 28. Du X et al (2012) Involvement of matrix metalloproteinase-2 in the development of renal interstitial fibrosis in mouse obstructive nephropathy. Lab Invest 92:1149–1160 29. Tao R et al (2010) Minocycline protects cardiac myocytes against simulated ischemia-reperfusion injury by inhibiting poly(ADPribose)polymerase-1. J Cardiovasc Pharmacol 56(6):659–668
123
ORIGINAL ARTICLE
Acute kidney injury after cardiac surgery: is minocycline protective? Ladan Golestaneh • Kathryn Lindsey • Pooja Malhotra • Faraj Kargoli • Emily Farkas • Hendrick Barner • Rizwan Qazi • Anna Schmidt • Michael Rauchman • Ziyad Al-Aly • Robert Johnson • Kevin Martin • Pierre Dagher • Allon Friedman • Tarek M. El-Achkar
Received: 6 August 2014 / Accepted: 13 October 2014 Ó Italian Society of Nephrology 2014
Abstract Background and objectives Acute kidney injury (AKI) after cardiac bypass surgery (CABG) is common and carries a significant association with morbidity and mortality. Since minocycline therapy attenuates kidney injury in animal models of AKI, we tested its effects in patients undergoing CABG. Design, setting, participants and measurements This is a randomized, double-blinded, placebo-controlled, multicenter study. We screened high risk patients who were scheduled to undergo CABG in two medical centers between Jan 2008 and June 2011. 40 patients were randomized and 19 patients in each group completed the study. Minocycline prophylaxis was given twice daily, at least for four doses prior to CABG. Primary outcome was defined as AKI [0.3 mg/dl increase in creatinine (Cr)] within 5 days after surgery. Daily serum Cr for 5 days, various clinical and hemodynamic measures and length of stay were recorded. Results The two groups had similar baseline and intraoperative characteristics. The primary outcome occurred in 52.6 % of patients in the minocycline group as compared to L. Golestaneh (&) P. Malhotra F. Kargoli Montefiore Medical Center, Albert Einstein Medical Center, 3411 Wayne Ave, Suite 5H, Bronx, NY 10467, USA e-mail: [email protected] K. Lindsey E. Farkas H. Barner R. Qazi A. Schmidt M. Rauchman R. Johnson K. Martin St Louis University Hospital, St Louis, MO 63110, USA Z. Al-Aly VA St Louis Health Care System, St Louis, MO 63106, USA P. Dagher A. Friedman T. M. El-Achkar IU Health University Hospital, Indianapolis, IN 46202, USA
36.8 % of patients in the placebo group (p = 0.51). Peak Cr was 1.6 ± 0.7 vs. 1.5 ± 0.7 mg/dl (p = 0.45) in minocycline and placebo groups, respectively. Death at 30 days occurred in 0 vs. 10.5 % in the minocycline and placebo groups, respectively (p = 0.48). There were no differences in post-operative length of stay, and cardiovascular events between the two groups. There was a trend towards lower diastolic pulmonary artery pressure [16.8 ± 4.7 vs. 20.7 ± 6.6 mmHg (p = 0.059)] and central venous pressure [11.8 ± 4.3 vs. 14.6 ± 5.6 mmHg (p = 0.13)] in the minocycline group compared to placebo on the first day after surgery. Conclusions Minocycline did not protect against AKI post-CABG. Keywords Acute kidney injury Cardiac surgery Minocycline
Introduction Acute kidney injury (AKI) after cardiac surgery remains a serious challenge [1]. The reported incidence of this complication depends on the definition used for AKI, but can reach as high as 30 %. The development of AKI is associated with increased in-hospital morbidity and mortality, and affects long-term survival [2]. Multiple interventions have attempted to prevent AKI after cardiac surgery but so far with disappointing results. Most of these interventions have centered on mitigating ischemia and inflammation, features of acute tubular necrosis (ATN), the presumed etiology of AKI in this setting. Risk factors for the development of postoperative AKI include patient related factors such as pre-existent kidney dysfunction, female gender, reduced left ventricular
123
J Nephrol
function, peripheral vascular disease, diabetes mellitus; and surgery related factors: including length of surgery, crossclamp time, off-pump versus on-pump, nonpulsatile flow, valvular surgery, hemolysis and hemodilution [4, 7, 9, 10]. The pathogenic basis of ATN was once thought to be a state of renal hypo-perfusion and resultant tissue ischemia [11]. However recently we have come to understand that ATN is downstream of earlier, highly complex events [12]. Injured endothelial cells initiate patterns of insult resulting in microvascular injury, leak and coagulation derangements. These processes lead to the activation of bonemarrow derived cells, various cytokines and interleukins with damage to endothelial and renal epithelial cells. Cardiac bypass is associated with activation of inflammatory cytokines and complement [13–17]. Tetracyclines possess a wide repertoire of anti-inflammatory and immunomodulatory actions in addition to their well-characterized antimicrobial effects [18, 19]. There is growing experimental evidence that tetracyclines such as minocycline, have a beneficial effect on cardiac myocytes and infarct size, the mechanism for which involves inhibition of free radical and cytokine production, interference with proteolysis, and modulation of matrix metalloproteinase activity [18–29]. Tetracyclines also possess potent antiapoptotic properties manifested by inhibition of caspase 1 and 3 expression and direct blockade of cytochrome c release from mitochondria [23, 24]. The potential benefits of minocycline and its high uptake by cardiac myocytes make it an ideal candidate for use in the setting of cardiac bypass surgery. This pilot study aimed at testing the hypothesis that pretreatment with minocycline of patients at high risk for AKI undergoing cardiac bypass surgery protects against developing acute kidney injury and improves overall cardiac outcomes.
Materials and methods Approval for this pilot study (ClinicalTrials.gov registration: NCT 00556491) was obtained from the St. Louis University and Montefiore Medical Center IRB committees. We adhered to the Declaration of Helsinki and informed consent was obtained from all subjects. Study design and patient population This study was designed as a double blinded, randomized, placebo controlled trial recruiting patients undergoing cardiac bypass surgery at Saint Louis University hospital (Missouri) and Montefiore Medical Center (New York). Patients were randomized to receive either minocycline capsules, 200 mg initially, then 100 mg every 12 h until
123
surgery, or placebo. Minocycline and placebo capsules looked identical and were provided by Triax. Both investigators and patients were blinded to the randomization, which was performed using permutated block method. The total number of doses, including the loading dose was at least 4 (i.e. 36 h prior to surgery minimum, last dose early morning on surgery day) to a maximum of 14 doses (7 days prior to surgery maximum). Patients were screened as soon as they were scheduled for bypass surgery. If eligible, and after informed consent was granted, they were randomized to either minocycline or placebo until surgery. Patients otherwise received standard medical and surgical care. Most of the parameters that were collected are part of the standard care of these types of surgical patients. Recruitment for the study started in January 2008 and finished in June 2011. The study population included all patients who were scheduled to undergo cardiac bypass surgery at both centers. They were patients with ischemic heart disease and/or valvular abnormalities who were scheduled to undergo open heart surgery (coronary artery bypass surgery with or without valvular repair/replacement). We did not target a specific age group, ethnicity or gender. However, the location and the disease entity of our study population excluded children a priori. To increase the event rate, we initially targeted patients with chronic kidney disease (CKD 2-4) calculated by MDRD formula. However, because of slow recruitment, and the fact that our urban centers delivered care to patients with a higher burden of chronic diseases such as diabetes, hypertension, and vascular disease (i.e. a high risk group), we removed the GFR limitation after year one of the study to include all ranges of GFR, excluding CKD 5. Our study did not include any subject population deemed as vulnerable. The inclusion and exclusion criteria are presented in Table 1. Verification that the patients took the medication was done by direct observation for inpatients. For patients who were randomized as outpatients, this was done by counting the remainder of the pills the day of surgery. Data collection Baseline demographic data and chronic medical illnesses for each patient enrolled in the study were recorded at randomization. Enrolled patient were monitored closely for signs and symptoms of any adverse effects of minocycline like allergic reactions, headaches, vertigo, dizziness and photosensitivity. In addition, major complications, and medication history were noted in detail. We monitored daily clinical and laboratory data needed for the study as described in the timeline, and labs were ordered in case they were missed by the primary team. We also collected hemodynamic data the 1st day when all patients had Swan-
J Nephrol Table 1 Baseline characteristics Minocycline (N = 19)
Placebo (N = 19)
Age (years)
64.1 ± 9.9
60.8 ± 11.5
p
0.34
White (%)
47.4
52.6
[0.9
African-American (%)
36.8
36.8
[0.9
Hispanic (%)
15.8
10.6
[0.9 [0.9
Male (%)
78.9
73.7
DM (%)
57.9
47.4
COPD (%) PVD (%)
15.8 31.6
15.8 36.8
CKD (%)
36.8
26.3
HTN (%)
94.7
94.7
[0.9
CHF (%)
47.4
42.1
[0.9 [0.9
0.75 [0.9 [0.9 0.73
EF B35 % (%)
26.3
26.3
Baseline Cr (mg/dl)
1.2 ± 0.5
1.1 ± 0.3
0.41
BMI
28.8 ± 3.6
32.4 ± 7.7
0.08
Albuminuria (%)
26.3
15.8
0.69
GFR (ml/min/1.73 m2)
71.9 ± 26.4
77.3 ± 27.4
0.54
Drug dose
6.4 ± 3.1
6.7 ± 3.3
0.78
Preop infection or antibiotics (%) Postop antibiotics (%)
0
10.0
0.15
68.4
47.4
0.90 0.08
Post op NSAIDS (%)
26.3
5.2
ACE I or ARB (%)
68.4
68.4
1.0
Diuretic use (%)
36.8
42.1
0.74
Statin use (%)
68.4
68.4
1.0
Beta blocker (%)
84.2
68.4
0.25
Calcium channel blocker (%)
26.3
42.1
0.30
Ganz catheter, detailed daily medication administration, fluid intake and output, ventilator use, length of stay in the ICU, and total hospital stay after surgery. Since 30 day mortality is a secondary objective in our study, a follow up phone call was made to patients to check their mortality status. Definitions In addition to the definitions discussed above we used the following other preset definitions: Baseline Cr was defined as the most recent Cr which typically was at the time of randomization. Our pre-specified primary outcome was the occurrence of AKI, defined as 0.3 mg/dl increase in Cr within any 48 h time period after surgery and up to a maximum of 5 days post-operatively. This definition is consistent with the acute kidney injury network (AKIN) criteria. Chronic kidney disease (CKD) was defined as an eGFR \60 ml/min as estimated by the 4-variable MDRD study equation. Secondary outcomes that were also reported were death up to 30 days post-op, cardiovascular
events, infections, and length of stay. We also examined cardiac hemodynamic parameters (cardiac index, systemic vascular resistance, pulmonary artery diastolic and central venous pressures) uniformly recorded at 6 a.m. on the first day after surgery from Swan-Ganz catheters. Data analysis Baseline and intra-operative variables of the control and intervention group were tabulated. A Fisher’s exact test was used to compare proportions for categorical variables and a t test was used to examine the difference in means for continuous data (reported as mean ± standard deviation). Univariate and regression analyses were performed as well to derive any significant associations with primary and secondary outcomes. Baseline variables included in this analysis were demographics (age, race), baseline kidney function and presence of albuminuria; chronic illnesses known to be risk factors for AKI after surgery such as diabetes mellitus (DM), hypertension (HTN), peripheral vascular disease (PVD), congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), body mass index (BMI); various medications by class such as diuretics, beta blockers, angiotensin converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARB) and statins. Intra-operative variables that were compared between the two groups included: surgery duration, crossclamp time, bypass (pump) time, lowest systolic pressure, day 1 lowest BP, day 1 cardiac output, day 1 diuretic use and day 1 statin use. Crude primary and secondary outcomes were compared using Fisher’s exact. A p value less than 0.05 were considered statistically significant. A t test was used to compare the difference in means for all hemodynamic measurements. Comparison of length of stay was done using t test after logarithmic transformation, since its distribution tends to be skewed. We also performed similar analysis on the whole population grouped into those patients who developed AKI vs. those who did not (no-AKI). This allowed us to determine whether certain risk factors were unevenly distributed in patients who developed AKI. Safety monitoring Adverse events were monitored until discharge. An independent Data Safety Monitoring Board (DSMB) had access to the unblended data. All pertinent safety data were reviewed from randomization until 1 week post-op and the serious adverse events (including death) up to 30 days post-op. The DSMB also performed quarterly reviews and advised the study investigators on any actions pertaining to the safety. All DSMB communications were reported to the IRB.
123
J Nephrol
Results Study flow, baseline demographics and intra-operative data We screened consecutive patients scheduled for cardiac surgery in two medical centers in St Louis and Bronx. The most common reason for not recruiting patients was the lack of timely notification to ensure there was enough time (at least 36 h) to give the drug prior to CABG. Another major factor was the emergent status of many surgeries. Other exclusion criteria are listed in Fig. 1. 40 patients were randomized to the study. Surgery was canceled in the case of 1 patient, and another patient withdrew himself from the study after randomization. 19 patients in each group underwent surgery and were included in the analysis. Table 1 shows the baseline characteristics of patients in the minocycline compared to the placebo groups. The two groups had similar baseline characteristics. The average age was 64.1 ± 9.9 in the minocycline group vs. 60.8 ± 11.5 in placebo (p = 0.34). Approximately half of the patients were of white ethnicity, predominantly of male gender. Diabetes was prevalent in 57.9 % of minocycline vs. 47.4 % of placebo groups (p = 0.75). Most of the population studied had HTN, and almost a third of the patients in each group had CKD. The prevalence of CHF in both groups was also high: 47.4 and 42.1 % in minocycline and placebo groups, respectively. Patients in the placebo group trended towards having a higher BMI, but this was not statistically significant (p = 0.08). There was no difference in blood pressure lowering medication intake by class (ACEI/ARB, beta blockers, calcium channel blockers, diuretics) or statin use between the two groups (Table 1). On baseline physical exam: 5.2 % of patient in the placebo
group had crackles, 10.5 % had edema. In the minocycline group: none of the patients had crackles and 15.8 % of patients had edema. On post-op day 1: 68.4 % of patients in the placebo group were given antibiotics (cefazolin and vancomycin); none of the patients were documented as having infection but one of the patients was classified as having a systemic inflammatory response syndrome (SIRS) likely from bypass. In the minocycline group: 47.4 % were given antibiotics consisting of cefazolin and vancomycin and none of the patients had documented infection or sepsis. After randomization, patients in the treatment group received an average of 6.4 ± 3.1 doses of minocycline, which was comparable to the number placebo doses given to the control group (6.7 ± 3.3, p = 0.78). The intraoperative data for patients in both study groups are presented in Table 2. The two groups had similar intraoperative characteristics. There was a trend towards a shorter cross-clamp time in the minocycline group compared to placebo (103.3 ± 39.3 vs. 141.1 ± 72.9, respectively; p = 0.06). The average lowest systolic BP was 79.9 ± 16.5 mmHg in the minocycline group vs. 88.7 ± 13.3 mmHg in placebo (p = 0.09). Duration of nonpulsatile flow in the placebo group was 172 ± 82 min, and 138 ± 57 min in the minocycline group. Primary and secondary outcomes Acute kidney injury, as defined by the AKIN criteria, occurred in 10 (58.3 %) patients in the minocycline group compared to 7 (36.8 %) patients in placebo (p = 0.51). There was no difference in the peak serum Cr between the two groups (Table 1). One patient required dialysis in each group. No patients died in the minocycline group compared to two deaths in the placebo group (p = 0.48). There was
Fig. 1 Serum creatinine levels after cardiac surgery; a comparing patient in the treatment vs. placebo groups and b comparing patient who developed AKI vs. those who did not
123
J Nephrol Table 2 Intra-op characteristics Minocycline (N = 19)
Table 3 Outcomes in each group Placebo (N = 19)
p
Minocycline N (%)
Placebo N (%)
p
Procedure duration (min)
355.5 ± 109.3
395.6 ± 127.2
0.32
AKI
10 (52.6)
7 (36.8)
0.51
Cross-clamp time (min)
103.3 ± 39.3
141.1 ± 72.9
0.06
Peak Cr (mg/dl)
1.6 ± 0.7
1.5 ± 0.7
0.45
Pump time (min)
138.0 ± 57.4
172.6 ± 85.1
0.15
Dialysis
1 (5.2)
1 (5.2)
1.0
79.9 ± 16.5
88.7 ± 13.3
0.09
Death (%)
0 (0)
2 (10.5)
0.48
Post-op hospital days
8.7 ± 3.8
9.8 ± 7.2
0.9
Post-op ICU days
4.8 ± 3.6
6.1 ± 7.1
0.75 0.45
Lowest systolic BP (mmHg) Urine output (ml)
991.5 ± 737.5
1,263.8 ± 888.3
0.34
no difference between the two groups in post-operative hospital and ICU stay (Table 2).
Ventilated [48 h
3 (15.8)
5 (26.3)
Infections
2 (10.5)
3 (15.8)
0.66
Stroke
2 (10.5)
0 (0)
0.48
Re-op
2 (10.5)
1 (5.6)
Minocycline (N = 19)
Placebo (N = 19)
p
92.9 ± 9.8
93.2 ± 12.6
0.94
Minocycline and hemodynamic parameters on day 1 post-operatively We examined whether minocycline treatment had an effect on hemodynamic parameters and cardiac pressures on day 1 after surgery. Table 2 shows that patients in the minocycline and placebo groups had comparable cardiac indices, systemic vascular resistances, and lowest systolic pressures. Patients in both groups received an equivalent amount of fluid, and had a comparable urine output. However, compared to placebo, patients in the minocycline group had a trend towards lower cardiac filling pressures as measured by the pulmonary artery diastolic (PAD) pressure (a surrogate for left ventricular end diastolic pressure): 16.8 ± 4.7 vs. 20.7 ± 6.6 mmHg in minocycline vs. placebo, respectively (p = 0.059). The central venous pressure (CVP) also trended lower in the minocycline group 11.8 ± 4.3 vs. 14.6 ± 5.9 mmHg in placebo (p = 0.13). There was no difference in these two pressures when we performed an analysis factoring BMI: BMI C30 vs. \30 [18.8 vs. 18.7 mmHg, respectively for PAD (p = 0.96); 13.3 vs. 13.1 mmHg, respectively for CVP (p = 0.92)]; thus uneven distribution of BMI between the treatment groups is unlikely to explain this finding. Characteristics of patients with AKI To better understand the characteristics and outcomes of patients who develop AKI, we grouped the patients in our study population based on whether they developed AKI, and examined all the variables (risk factors, intra-op events, and outcomes). Table 3 shows only the significant variables that were unevenly distributed with each group. Patients with AKI were more likely to be obese, and have peripheral vascular disease. They also tend to have longer surgeries, and develop more intra-operative hypotension. Patients with AKI stayed longer in the ICU and in the hospital after surgeries. Figure 1a shows the average serum Cr between the two groups. Figure 1b shows that AKI is most commonly
Lowest systolic BP (mmHg)
[0.9
Fluid Intake (ml)
2,923.6 ± 1,197.5
3,134.5 ± 946.4
0.57
Urine Output (ml)
2,031.4 ± 1,179.6
2,355.8 ± 931.0
0.40
Cardiac Index
2.8 ± 0.7
3.0 ± 0.9
0.53
SVR
1,051.6 ± 342.3
989.6 ± 355.1
0.61
PAD
16.8 ± 4.7
20.7 ± 6.6
0.059
CVP
11.8 ± 4.3
14.6 ± 5.9
0.13
Vasopressor use (%)
58.8
50.0
0.74
Ventilator (%)
52.9
66.7
0.73
BUN
19.2 ± 7.4
16.7 ± 3.9
0.21
Cr
1.3 ± 0.5
1.1 ± 0.3
0.22
detected at day 2 post-operatively and the serum Cr on average peaks at day 3. Using regression analysis the following baseline parameters were significantly associated with developing AKI post-op: baseline PVD, higher BMI, pre-op CKD as marked by abnormal creatinine, and pre-op age. Minocycline safety Minocycline was very well tolerated by all patients. Only one serious adverse event occurred prior to surgery in a patient on chronic anticoagulation, where treatment with minocycline may have contributed to him being supratherapeutic on Coumadin, requiring admission for anemia and gastro-intestinal bleeding, that was treated conservatively. There was no difference in serious adverse events post-operatively, as shown in Table 3.
Discussion In this pilot study, we examined the effect of minocycline prophylaxis on developing AKI after cardiac surgery with
123
J Nephrol
cardio-pulmonary bypass. Our study was a prospective study with comprehensive data gathering including intraop and post-op parameters that are seldom gathered in other studies. We examined secondary outcomes, including risk factors for AKI, using the specific data points derived. Although our pilot study did not find an effect of minocycline on incidence of AKI after CABG, the findings could help design a large and adequately powered trial. The therapeutic effects of tetracyclines on cardiac muscle post-ischemia are time sensitive: it is important to block proteases that contribute to pathologic cell remodeling and not block those that help surrounding tissue recover [20–28]. For example: post-ischemia minocycline treatment was protective in a rodent model of atherosclerotic cardiac ischemia, as was administration of minocycline prior to 36 h before the renal clamping in an AKI murine model [27, 28]. The timing of minocycline administration is an open question with regards to future studies. In any clinical study of AKI, there is debate about the applicability of animal models. The minocycline data is no different. While administration of minocycline pre and/ or post ischemia was shown to be protective in rodent clamp models of ischemic ATN, whether this agent protects against cardiac surgery specific renal injury is unclear [27, 28]. In our study minocycline treatment appeared to be associated with lower cardiac filling pressures despite similar fluid balance and cardiac indices. This suggests that minocycline induces a state of improved cardiac compliance, which could be beneficial for cardiovascular and pulmonary outcomes. The effect of minocycline on cardiac compliance does have a biological explanation. Studies in rodents showed that minocycline rapidly accumulates in the myocardium after administration [18, 27, 28]. Minocycline also had protective effects on the myocardium postinfarction, and inhibited oxidative stress [28]. Tao et al. [29] showed more recently that minocycline protects cardiac myocytes possibly through inhibition of poly(ADPribose) polymerase-1 (PARP-1). Extensive PARP-1 activation can lead to cell death through bio-energetic failure and depletion of ATP. Pharmacologic and genetic inhibition of PARP-1 improved cardiac contractile function in an experimental model of heart failure [29]. Therefore, we propose that minocycline improves cardiac compliance post CABG by its protective effects on the myocardium and possibly though preserving energy stores by PARP-1 inhibition. This finding underscores the importance of performing a larger trial to examine the effect of minocycline prophylaxis on cardiovascular, pulmonary and length of stay outcomes. Significant factors associated with developing AKI in our population were the presence of peripheral vascular
123
disease, age and presence of CKD at baseline. This is consistent with previously published data [2–11]. Moreover, longer surgeries and intraoperative hypotension were also more common in patients with AKI. Examining intraoperative data is essential to determining the risks of AKI in patients undergoing cardiac bypass. The association of AKI with length of stay is in accordance with previous observational data from Chertow et al. [2]. The incidence of AKI in our study population (45.6 %) was higher than reported by previous studies. This could be explained by the use of a very sensitive definition (AKIN stage 1) and the recruitment of a high risk population, where diabetes, HTN, PVD and CKD were highly prevalent. All surgeries by design were performed using bypass, which carries a higher risk for AKI than off pump cardiac surgery. Our study population included a significant proportion of African-American and Hispanic participants. This could also influence the incidence of AKI, but at the same time improve the validity of the findings. With the advent of new diagnostic biomarkers the possibility of detecting AKI earlier and more accurately than what is offered by serum creatinine, may be realized. NGAL has been shown to be of value in the early diagnosis of AKI in the post cardiac surgery setting; however consensus guidelines use serum creatinine for AKI definition. Future projects will incorporate a diverse panel of biomarkers, once they are validated. A major strength of our study is in its design as a randomized, double blinded, placebo controlled study. Our study is unique in providing detailed intraoperative variables, and post-operative data such as hemodynamic measurement. Furthermore, our study recruited patients from two different centers located in two states, and recruited a diverse population, which improved its validity. The major limitation of this study is its sample size, which precludes major conclusions on the primary outcome. However, our study can be used as a model to design future studies to administer prophylactic interventions in patients undergoing cardiac surgery. In conclusion, in our pilot study, minocycline prophylaxis prior to cardiac surgery with bypass did not reduce the occurrence of AKI, but could have a beneficial effect on cardiac compliance. These findings provide a rationale and a roadmap to perform a large trial that investigates not only renal, but also cardiovascular, pulmonary and length of stay endpoints. Conflict of interest This study was supported by a Norman Coplon Grant from Satellite Healthcare, and by a limited grant from Triax Pharmaceuticals (makers of generic form of minocycline).
J Nephrol
References 1. Mehta RL (2005) Acute renal failure and cardiac surgery: marching in place or moving ahead? J Am Soc Nephrol 16(1):12–14 2. Chertow GM et al (1998) Independent association between acute renal failure and mortality following cardiac surgery. Am J Med 104(4):343–348 3. Fortescue EB, Bates DW, Chertow GM (2000) Predicting acute renal failure after coronary bypass surgery: cross-validation of two risk-stratification algorithms. Kidney Int 57(6):2594–2602 4. Slogoff S et al (1990) Role of perfusion pressure and flow in major organ dysfunction after cardiopulmonary bypass. Ann Thorac Surg 50(6):911–918 5. Thakar CV et al (2005) A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol 16(1):162–168 6. Thakar CV et al (2003) ARF after open-heart surgery: influence of gender and race. Am J Kidney Dis 41(4):742–751 7. Tuttle KR et al (2003) Predictors of ARF after cardiac surgical procedures. Am J Kidney Dis 41(1):76–83 8. Chertow GM et al (1997) Preoperative renal risk stratification. Circulation 95(4):878–884 9. Abramov D et al (2003) The influence of cardiopulmonary bypass flow characteristics on the clinical outcome of 1820 coronary bypass patients. Can J Cardiol 19(3):237–243 10. Fischer UM et al (2002) Impact of cardiopulmonary bypass management on postcardiac surgery renal function. Perfusion 17(6):401–406 11. Rosner MH, Okusa MD (2006) Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol 1:19–32 12. Devarajan P (2006) Update on mechanisms of ischemic acute kidney injury. J Am Soc Nephrol 17(6):1503–1520 13. Molitoris BA (2003) Transitioning to therapy in ischemic acute renal failure. J Am Soc Nephrol 14(1):265–267 14. Molitoris BA, Sutton TA (2004) Endothelial injury and dysfunction: role in the extension phase of acute renal failure. Kidney Int 66(2):496–499 15. Okusa MD (2002) The inflammatory cascade in acute ischemic renal failure. Nephron 90(2):133–138 16. Sheridan AM, Bonventre JV (2000) Cell biology and molecular mechanisms of injury in ischemic acute renal failure. Curr Opin Nephrol Hypertens 9(4):427–434
17. Sutton TA, Fisher CJ, Molitoris BA (2002) Microvascular endothelial injury and dysfunction during ischemic acute renal failure. Kidney Int 62(5):1539–1549 18. Griffin MO et al (2010) Tetracyclines: a pleitropic family of compounds with promising therapeutic properties. Review of the literature. Am J Physiol Cell Physiol 299:C539–C548 19. Nieman GF, Zerler BR (2001) A role for the anti-inflammatory properties of tetracyclines in the prevention of acute lung injury. Curr Med Chem 8(3):317–325 20. Arvin KL et al (2002) Minocycline markedly protects the neonatal brain against hypoxic-ischemic injury. Ann Neurol 52(1):54–61 21. Cheung PY et al (2000) Matrix metalloproteinase-2 contributes to ischemia-reperfusion injury in the heart. Circulation 101(15):1833–1839 22. Sloan R et al (2012) Mitochondrial permeability transition in the diabetic heart: contributions of thiol redox state and mitochondrial calcium to augmented reperfusion injury. J Mol Cell Card 52:1009–1018 23. Tikka T et al (2001) Tetracycline derivatives and ceftriaxone, a cephalosporin antibiotic, protect neurons against apoptosis induced by ionizing radiation. J Neurochem 78(6):1409–1414 24. Power C et al (2003) Intracerebral hemorrhage induces macrophage activation and matrix metalloproteinases. Ann Neurol 53(6):731–742 25. Chen M et al (2000) Minocycline inhibits caspase-1 and caspase3 expression and delays mortality in a transgenic mouse model of Huntington disease. Nat Med 6(7):797–801 26. Zhu S et al (2002) Minocycline inhibits cytochrome c release and delays progression of amyotrophic lateral sclerosis in mice. Nature 417(6884):74–78 27. Xia D et al (2011) Adminsitration of minocycline ameliorates damage in a renal ischemia/reperfusion injury model. Clin Invest Med 34(2):E55–E63 28. Du X et al (2012) Involvement of matrix metalloproteinase-2 in the development of renal interstitial fibrosis in mouse obstructive nephropathy. Lab Invest 92:1149–1160 29. Tao R et al (2010) Minocycline protects cardiac myocytes against simulated ischemia-reperfusion injury by inhibiting poly(ADPribose)polymerase-1. J Cardiovasc Pharmacol 56(6):659–668
123
