http://orcid.org/0000-0001-7688-8155
Abstract
Background: Acute kidney injury (AKI) is a postoperative complication associated with significant morbidity and mortality. The incidence and risks factors for AKI after cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (CRS-HIPEC) have not been fully studied. The purpose of this study was to identify perioperative risk factors predictive of AKI after CRS-HIPEC.
Methods: This retrospective study collected demographic, tumour-related, intraoperative and postoperative data from 475 patients who underwent CRS-HIPECs. AKI was defined using the acute kidney injury network criteria and calculated on postoperative days 1, 2, 3, 7 and day of hospital discharge. We conducted univariate and multivariate analyses to assess the association between variables of interest and AKI. A p value of <0.05 was considered statistically significant.
Results: The incidence of AKI was 21.3%. The multivariate analysis identified six predictor factors independently associated with the development of AKI (OR: [95%CI]); age: 1.16 (1.05–1.29, p < 0.005), BMI (overweight: 1.97 [1.00–3.88], p = 0.05) and obesity: 2.88 (1.47–5.63), p < 0.002)), preoperative pregabalin: 3.04 (1.71–5.39, p < 0.037), platinum-based infusion: 3.04 (1.71–5.39, p < 0.001) and EBL: 1.77 (1.27–2.47, p < 0.001). Splenectomy had a protective effect (OR: 0.44 (0.25–0.76, p < 0.003).
Conclusions: Our study demonstrates that the incidence of AKI is high. While other studies have reported that AKI is associated with platinum-based infusion, age and obesity, we report for the first time a negative association between pregabalin use and AKI. More studies are needed to confirm our results.
Introduction
Cytoreductive surgery (CRS) with hyperthermic intraperitoneal chemotherapy (HIPEC) is a major abdominal oncological surgery that combines extensive tumour debulking, peritonectomy, multiple visceral resections and the delivery of hyperthermic chemotherapy [Citation1]. In recent years, the number of centres worldwide that are performing CRS-HIPEC procedures has significantly increased due to improvements in surgical techniques, expansion in the indications and a better understanding of the perioperative care of patients undergoing this procedure. However, the rate of postoperative complications remains high, and they significantly impact the survival of patients [Citation2–4]. Only a few large studies have reported on the results of perioperative complications such as acute kidney injury (AKI) [Citation5].
Acute kidney injury (AKI) is a postoperative complication associated with significant morbidity and mortality [Citation6]. It has been proposed that the hyperthermic phase of the surgery resembles a model of septic shock [Citation7]. Therefore, several mechanisms including ischaemia–reperfusion secondary to episodes of arterial hypotension, fluid redistribution, the direct effects of nephrotoxic drugs and inflammation have been proposed as some of the pathophysiological mechanisms responsible for perioperative AKI. Based on these premises, it has been speculated that patients undergoing CRS-HIPEC surgery have an increased risk of developing AKI because of frequent episodes of intraoperative hemodynamic instability, major fluid shifts, the administration of drugs such as cisplatin and an exaggerated inflammatory response [Citation8].
To date, the actual incidence and risk factors associated with the development of postoperative AKI after CRS-HIPEC surgery remain poorly studied. The reported incidence ranges between 2% and 22%; however, most of the studies reporting this complication had limited sample sizes and have used different systems to define renal insufficiency [Citation9–12]. Hence, we conducted a retrospective study with the aim of identifying risk factors of AKI in adult patients who underwent HIPEC surgery in a large tertiary cancer care centre. We considered that the identification of modifiable and non-modifiable risk factors for AKI after CRS-HIPEC is needed since this complication appears to be one of the strongest predictors of overall morbidity [Citation13].
Materials and methods
After obtaining MD Anderson Cancer Center Institutional Review Board approval (#PA14-1087), we conducted a retrospective cohort study that included data from patients who underwent CRS-HIPEC between 1 January 2006 and 19 February 2016. We included patients who were 18 years or older who had open or minimally invasive CRS-HIPECs. Patients with incomplete perioperative data and those with preoperative renal replacement therapy (dialysis) were excluded. The following demographic, tumour-related, intraoperative and postoperative variables were collected from electronic medical records: age, gender, American Society of Anesthesiologists (ASA) physical status, body mass index (BMI), preoperative haemoglobin, comorbidities including systemic hypertension (HTN), diabetes mellitus (DM) coronary artery disease (CAD), congestive heart failure, chronic obstructive pulmonary disease, cerebrovascular accident, chronic renal insufficiency (creatinine >2 mg/dL), perioperative medications, history of neoadjuvant chemotherapy, peritoneal carcinomatosis index (PCI), completeness of cytoreduction score (CC), splenectomy, tumour histology and grade, type and duration of chemotherapy administration, fluid administration (type and volume) and urinary output. Preoperative serum creatinine followed by postoperative creatinine at days 1, 2, 3 and 7 and at hospital discharge were collected.
Outcomes
The primary outcome of this study was AKI incidence within 72 h after surgery, on postoperative day 7 and the time of hospital discharge. We estimated the incidence of AKI using AKIN criteria as defined in previous studies [Citation14,Citation15].
Perioperative care
Preoperative oral tramadol (300 mg), celecoxib (400 mg) or pregabalin (75 mg) was typically administered within two hours of anaesthesia induction according to anaesthesiologist clinical judgement. All patients had general anaesthesia using a combination of opioids, volatile anaesthetic (i.e. desflurane) in oxygen or total intravenous anaesthesia (propofol) and muscle relaxation with a non-depolarising or a depolarising muscle relaxant. Intraoperative fluid therapy consisted of a combination of crystalloids (Plasmalyte®, Baxter International, Deerfield, IL) and colloids (6% hetastarch, Hextend®, BioTime, Inc., Berkeley, CA; or albumin 5%, Buminate®, Baxter, San Juan, Puerto Rico) that was administered according to the attending anaesthesiologist’s clinical judgement. Ureteral stents were placed before surgical incision at discretion of primary surgery. Intraperitoneal chemotherapy was administered according to tumour type and surgeon’s choice. Perioperative blood product transfusions were given to maintain a haemoglobin (Hb) concentration between 8 and 10 g/dL. Postoperative analgesia consisted of either patient-controlled epidural analgesia and/or intravenous patient-controlled analgesia.
Statistical analysis
Descriptive statistics were used to summarise the demographic and perioperative clinical data. Comparisons were made between patients with and without AKI across all candidate predictors. Both demographic and perioperative variables were used to identify candidate predictors of AKI. As a method of screening predictors, logistic regression in a univariable setting occurred first. Continuous predictors were assessed to ensure whether linear functional forms were tenable. The strength of the relationship between AKI and all potential predictors that meet a threshold of p < 0.25 was considered for inclusion into a full model [Citation16]. Furthermore, once a full model was identified, model selection through backwards elimination was used to derive a reduced model. The final multivariable model was derived using backwards elimination with the probability of removal set at 0.05 to provide a reduced model. The Hosmer–Lemeshow goodness-of-fit test along with the area under the ROC curve (AUC) was provided as measures of goodness-of-fit and model performance, respectively.
Results
The average (standard deviation) patient age was 51 (12.65) years (). There were more female (55.1%) than male (44.8%) patients. The majority of participants were overweight (35.8%) or obese (35.0%), and over 80% of patients had an American Society of Anesthesiologists (ASA) score of 3. Diabetes mellitus (37%) and chronic pulmonary disease (15%) were the most common comorbidities. A small percentage (3.1%) of the patients were taking cardiovascular medications at the time of surgery. Most patients (71.4%) had appendiceal cancer. Other tumour diagnoses included: colorectal (10.3%), mesothelioma (8.8%), desmoplastic round cell tumour (7.0%) and other cancers (2.5%). Forty-three percent of patients had moderate to high tumour grades, and 44.8% of patients received neoadjuvant chemotherapy. At the time of surgery, only 3.5% of the patients had extra abdominal disease.
Table 1. Patient demographics.
Acute kidney injury
Of the 475 patients in our study, 101 (21.3%) were identified as having an AKI stage 1 or higher postoperatively. As shown in , patients who developed postoperative AKI were older (53.9 ± 12.17 years) than those (50.1 ± 12.6 years) without AKI. AKI was more frequent in male (57.4%) and overweight/obese (84.1%) patients, as well as in those with preoperative moderate-to-severe chronic kidney disease (). Other demographic and preoperative variables were not statistically different in patients with and without AKI (). As shown in , a higher percentage of patients with AKI had mesothelioma (17.82%) and received cisplatin intraoperatively (28.71%). Interestingly, AKI was statistically less frequent in patients who had splenectomies (). The amount of vasopressors administered during surgery was statistically significant for ephedrine but not for phenylephrine and dopamine (). Patients with AKI received a higher cumulative dose of ephedrine (32.61 mg ±35.19) than those without AKI (23.28 mg ±17.75). Patients with AKI received preoperative pregabalin in a significantly larger proportion, had longer duration of surgery, were treated with a larger amount of crystalloids, had larger estimated blood loss and were transfused in a larger proportion than those without AKI (). The rate of ureteral stent placement was significantly higher in patients with AKI (74.26%) than dose without AKI (62.03%). Also shown in , patients without AKI received more perioperative NSAIDs in a larger percentage than those with AKI. The analysis demonstrated that the prevalence of AKI on postoperative day 7 and day hospital discharge was 5.66% and 4.21%, respectively. None of the patients who developed AKI needed renal replacement therapy.
Table 2. Tumour and intraoperative chemotherapy variables.
Table 3. Perioperative medications and intraoperative fluid therapy.
Univariate and multivariate analysis
We considered that the relationship between AKI and all potential predictors was strong if it was lower than a threshold of p < 0.25. The following 12 covariates: age, gender, body mass index, splenectomy, extra abdominal disease, tumour grade and chemotherapy, NSAIDs use, estimated blood loss, postoperative blood products, anaesthesia duration and total crystalloids administration were considered strong. The distribution of estimated blood loss was highly skewed to the right; therefore, the data were transformed using the natural logarithm. Body mass index was entered into the model as a categorical variable with 3 levels: normal (BMI < 25 kg/m2), overweight (25 kg/m2 < BMI < 30 kg/m2) and obese (>30 kg/m2). After predictor screening was conducted, three covariates were eliminated due to their extent of missing data. Moreover, the following three covariates were strongly correlated: anaesthesia duration, total crystalloids, estimated blood loss and ureteral stent placement. Thus, we created two multivariate models. The first multivariable model contained the following six predictors (OR: [95%CI]) of postoperative AKI: age: 1.16 (1.05–1.29), BMI (overweight: 1.97 [1.00–3.88]) and obesity: 2.88 (1.47–5.63)), preoperative pregabalin: 3.04 (1.71–5.39), platinum-based infusion: 3.04 (1.71–5.39) and EBL: 1.77 (1.27–2.47). Patients who required a splenectomy had a significantly lower risk of AKI [OR: 0.44 (0.25–0.76)] (). Our analysis demonstrated that preoperative pregabalin and ureteral stents were significantly associated. Of patients receiving preoperative pregabalin, 92.4% had ureteral stents (p < 0.001). From the univariable analysis, the association between AKI and ureteral stent was slightly more significant than pregabalin use. Therefore, ureteral stents were chosen over preoperative pregabalin to remain in the second model because it showed a slightly stronger association with AKI. In this second multivariate analysis, age [OR: 1.15 (1.04, 1.28)], BMI [obesity, OR: 2.93 (1.49, 5.74)], ureteral stents [OR: 1.92 (1.13, 3.28)] and platinum-based chemotherapy [OR: 3.02 (1.70, 5.34)] were independent risk factors for AKI, while splenectomy [OR: 0.42 (0.24, 0.73)] showed a protective effect ().
Table 4. Results of multivariable logistic regression – Model 1.
Table 5. Results of multivariable logistic regression – Model 2.
Discussion
The incidence of AKI in this large cohort of patients who had undergone CRS-HIPEC was 21%. Reported rates of AKI after CRS-HIPEC have ranged between 2% and 22% [Citation9–13,Citation17,Citation18]. For example, Hamilton et al. found that 7.1% of the 42 patients in their study developed acute renal failure after CRS-HIPEC. Unfortunately in that study, the authors did not provide the definition of AKI [Citation17]. Using a larger sample size (n = 141) and the AKIN criteria, Arjona-Sanchez et al. reported an almost identical rate of AKI to ours. They found that 22% of patients had AKI after CRS-HIPEC [Citation18]. Previous studies reported that the incidence of severe AKI is unacceptably high (1.3%–5.7%) [Citation9,Citation10,Citation19].
Our study also demonstrated that the development of AKI was significantly associated with the type of chemotherapy infused, EBL, administration of pregabalin, BMI and age. Patients who received cisplatin or oxaliplatin as the HIPEC agent had a higher risk of AKI compared to those who received mitomycin C. While this finding is similar to what has been described in previous studies [Citation11,Citation20], the extent to which this association may be wholly ascribed to the direct nephrotoxic effects of the platinum agents has been the subject of controversy [Citation8,Citation21]. The basis of the disagreement is that plasma levels of HIPEC agents including cisplatin have been demonstrated to be well below the cytotoxic threshold and may therefore not be sufficient to explain the nephrotoxicity observed [Citation22].
Adequate hydration has been noted to be essential to the prevention of renal toxicity during CRS-HIPEC [Citation23]. In our study, there was notable lack of association between intraoperative fluids or urine output, use of and the development of AKI. This may be explained by the fact that our institution follows the European Society of Clinical Pharmacy Special Interest Group on Cancer Care recommendation that suggests “not to administer platinum compounds to patients before objective evidence of euvolemia is present” [Citation24]. However, this cannot explain why the administration of platinum-based agents was an independent risk factor of AKI in our study.
It has been proposed that other factors such as hypoperfusion as a result of intra-abdominal hypertension or systemic hypotension may also play a role in the development of AKI [Citation8]. It is possible to speculate that a higher EBL could have resulted in longer periods/episodes of hypotension, while a higher BMI would be associated with increased intra-abdominal pressure. Both hypotension and obesity would result in organ hypoperfusion and ischaemia. Our study found that the total amount of phenylephrine and dopamine used during surgery was not statistically different in patients with and without AKI. Contrarily, the amount of ephedrine used was significantly higher in patients with AKI. We believe that 9 mg of difference is not a clinically relevant difference. Thereby we did not include the use of vasopressors in the multivariate analysis.
Furthermore, obese patients exhibit a higher risk of AKI because of hyperfiltration, increased glomerular capillary wall tension and podocyte stress [Citation25]. Blood transfusions are associated with postoperative AKI in HIPEC surgery [Citation26]. Our study found that the rate of postoperative blood transfusions was higher in patients with AKI. This could be the result of high EBLs and a larger intraoperative fluid administration in that group of patients in whom postoperative haemodilution could have triggered a more frequent use of blood transfusions.
Our study suggests that the placement of ureteral stents is associated with postoperative AKI. The literature is scarce on the impact of intraoperative ureteral stent placement for non-urological procedures and AKI. Arjona-Sanchez et al. reported a similar association to our study. The authors speculated that the combination of residual hydronephrosis from the ureteral stents, mucosal oedema, clots and renal hypoperfusion could explain the increased risk of renal injury [Citation27]. Reflex anuria has been described as a cause of renal injury [Citation28]. Our experience suggests that in patients with significant intraoperative oliguria or anuria, removing the ureteral stents improves urinary output.
Age and diabetes-associated declines in glomerular filtration rate (GFR) and renal blood flow have been recognised for decades [Citation29–31]. For instance, arterial thickening and glomerulopathic changes have also been observed in the kidneys of older donors and diabetic patients [Citation32,Citation33]. Although it was not surprising that age was significantly associated with the development AKI in our study population, we did not find that diabetes mellitus was an independent predictor of this complication. It is possible that in diabetic patients, our anaesthesiologist may have adopted a more aggressive approach in terms of fluid resuscitation and avoidance of cytotoxic drugs (i.e. NSAIDs).
The observed association between pregabalin administration and the development of AKI is an unexpected finding. It is likely that some unexplored factors may be able to explain these associations. However, recent evidence indicates that gabapentinoids can induce muscle damage [Citation34,Citation35]. Therefore, it is possible to hypothesise that, in the context of hyperthermia and cisplatin administration, even low circulating concentrations of myoglobin can be toxic to the kidneys. It is worth considering that a significant interaction was found between pregabalin use and ureteral stent placement during surgery. Therefore, in our second multivariate model, ureteral stent remained as an independent predictor of AKI.
Lastly, we observed that splenectomised patients had a lower risk of developing AKI. The impact of splenectomy on distant organ injury appears to be organ dependent. While splenectomy is an independent risk factor of postoperative complications including early postsurgical infections and lung injury, it appears to have a protective effect on ischaemic damage to the brain and liver [Citation36–39]. In support of our findings, splenectomy has been shown to have a significant beneficial effect in an animal model of kidney ischaemia–reperfusion injury [Citation40].
Our study has significant problems related to its retrospective design. Significant selection, information and detection biases limit the interpretation of our results. First, we did not include in our analysis data such as disease severity or intraoperative and early postoperative haemodynamics. Episodes of systemic hypotension are known to significantly impact the development of AKI [Citation6]. Second, in our institution, more frequent serum creatinine determinations are typically made in patients at risk for AKI, which is a source of detection bias. Third, we used creatinine concentrations to estimate AKI. It is well known that subtle deteriorations in kidney function are not detected by creatinine. Furthermore, we did not adjust the measured creatinine concentrations to changes in total body water [Citation41]. Therefore, it is possible that our findings represent an underestimation of the actual incidence of early postoperative AKI. Furthermore, we did not use the RIFLE criteria, which also might have been a source of underestimation of AKI [Citation18]. Unfortunately, not all patients had accurate long-term follow-up to use RIFLE criteria. Lastly, our analysis did not include data on early postoperative complications that are known predictors of AKI, such as sepsis or massive haemorrhage.
In conclusion, age, elevated BMI, the administration of platinum-based agents, the use of pregabalin, ureteral stents and major blood loss are independent risk factors for the development of AKI. Randomised controlled trials should be designed studies on intra- and postoperative management of HIPEC patients should be designed to test interventions targeted to maintain systemic perfusion and organ support.
Disclosure statement
The authors have not conflict of interest to disclose.
Additional information
Funding
References
- Sugarbaker PH. (1995). Peritonectomy procedures. Ann Surg 221:29–42.
- Lee L, Alie-Cusson F, Dube P, et al. (2017). Postoperative complications affect long-term outcomes after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for colorectal peritoneal carcinomatosis. J Surg Oncol 116:236–24.
- Bhatt A, Sheshadri D, Chandan G, et al. (2017). Outcomes of cytoreductive surgery and HIPEC for pseudomyxoma peritonei of appendiceal origin from two Indian centers: a preliminary five-year experience. J BUON 22:251–7.
- Schneider MA, Eshmuminov D, Lehmann K. (2017). Major postoperative complications are a risk factor for impaired survival after CRS/HIPEC. Ann Surg Oncol 24:2224–32.
- Owusu-Agyemang P, Cata JP, Fournier KF, et al. (2016). Evaluating the impact of total intravenous anesthesia on the clinical outcomes and perioperative NLR and PLR profiles of patients undergoing cytoreductive surgery with hyperthermic intraperitoneal chemotherapy. Ann Surg Oncol 23:2419–29.
- Walsh M, Devereaux PJ, Garg AX, et al. (2013). Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: toward an empirical definition of hypotension. Anesthesiology 119:507–15.
- Coccolini F, Corbella D, Finazzi P, et al. (2016). Time course of cytokines, hemodynamic and metabolic parameters during hyperthermic intraperitoneal chemotherapy. Minerva Anestesiol 82:310–9.
- Ceresoli M, Coccolini F, Ansaloni L. (2016). HIPEC and nephrotoxicity: a cisplatin induced effect? Eur J Surg Oncol 42:909–10.
- Kusamura S, Baratti D, Younan R, et al. (2007). Impact of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy on systemic toxicity. Ann Surg Oncol 14:2550–8.
- Glehen O, Osinsky D, Cotte E, et al. (2003). Intraperitoneal chemohyperthermia using a closed abdominal procedure and cytoreductive surgery for the treatment of peritoneal carcinomatosis: morbidity and mortality analysis of 216 consecutive procedures. Ann Surg Oncol 10:863–9.
- Bouhadjari N, Gabato W, Calabrese D, et al. (2016). Hyperthermic intraperitoneal chemotherapy with cisplatin: amifostine prevents acute severe renal impairment. Eur J Surg Oncol 42:219–23.
- Saxena A, Valle SJ, Liauw W, et al. (2017). Limited synchronous hepatic resection does not compromise peri-operative outcomes or survival after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. J Surg Oncol 115:417–24.
- Plackett TP, Ton-That HH, Mosier MJ, et al. (2017). Physiologic response to HIPEC: sifting through perturbation to identify markers of complications. J Am Osteopath Assoc 117:16–23.
- Moon T, Tsai JY, Vachhani S, et al. (2016). The use of intraoperative dexmedetomidine is not associated with a reduction in acute kidney injury after lung cancer surgery. J Cardiothorac Vasc Anesth 30:51–5.
- Ishikawa S, Griesdale DE, Lohser J. (2012). Acute kidney injury after lung resection surgery: incidence and perioperative risk factors. Anesth Analg 114:1256–62.
- Hosmer DW, Lemeshow S. (2014). Logistic regression models for the analysis of correlated data. In: Hosmer DWL, LemeshowS, eds. Applied logistic regression. 3rd ed. Canada: John Wiley & Sons, 213.
- Hamilton TD, Taylor EL, Cannell AJ, et al. (2016). Impact of major complications on patients' quality of life after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. Ann Surg Oncol 23:2946–52.
- Arjona-Sanchez A, Cadenas-Febres A, Cabrera-Bermon J, et al. (2016). Assessment of RIFLE and AKIN criteria to define acute renal dysfunction for HIPEC procedures for ovarian and non ovarian peritoneal malignances. Eur J Surg Oncol 42:869–76.
- Verwaal VJ, van Tinteren H, Ruth SV, et al. (2004). Toxicity of cytoreductive surgery and hyperthermic intra-peritoneal chemotherapy. J Surg Oncol 85:61–7.
- Hakeam HA, Breakiet M, Azzam A, et al. (2014). The incidence of cisplatin nephrotoxicity post hyperthermic intraperitoneal chemotherapy (HIPEC) and cytoreductive surgery. Ren Fail 36:1486–91.
- Bouhadjari N, Gabato W, Calabrese D, et al. (2016). Reply letter to: HIPEC and nephrotoxicity: a cisplatin induced effect? Eur J Surg Oncol 42:907–8.
- Ansaloni L, Coccolini F, Morosi L, et al. (2015). Pharmacokinetics of concomitant cisplatin and paclitaxel administered by hyperthermic intraperitoneal chemotherapy to patients with peritoneal carcinomatosis from epithelial ovarian cancer. Br J Cancer 112:306–12.
- Raspe C, Flother L, Schneider R, et al. (2016). Best practice for perioperative management of patients with cytoreductive surgery and HIPEC. Eur J Surg Oncol 43:1013–27.
- Launay-Vacher V, Rey JB, Isnard-Bagnis C, et al. (2008). Prevention of cisplatin nephrotoxicity: state of the art and recommendations from the European Society of Clinical Pharmacy Special Interest Group on Cancer Care. Cancer Chemother Pharmacol 61:903–9.
- Wickman C, Kramer H. (2013). Obesity and kidney disease: potential mechanisms. Semin Nephrol 33:14–22.
- Sin EIL, Chia CS, Tan GHC, et al. (2017). Acute kidney injury in ovarian cancer patients undergoing cytoreductive surgery and hyperthermic intra-peritoneal chemotherapy. Int J Hyperthermia 33:690–5.
- Pathak RA, Taylor AS, Alford S, et al. (2015). Urologic-induced complications of prophylactic ureteral localization stent placement for colorectal surgery cases. J Laparoendosc Adv Surg Tech A25:966–70.
- Adediran S, Dhakarwal P. (2014). Reflex anuria: a rare cause of acute kidney injury. J Community Hosp Intern Med Perspect 4:23423.
- Weinstein JR, Anderson S. (2010). The aging kidney: physiological changes. Adv Chronic Kidney Dis 17:302–7.
- Sobamowo H, Prabhakar SS. (2017). The kidney in aging: physiological changes and pathological implications. Prog Mol Biol Transl Sci 146:303–40.
- Cata JP, Zafereo M, Villarreal J, et al. (2015). Intraoperative opioids use for laryngeal squamous cell carcinoma surgery and recurrence: a retrospective study. J Clin Anesth 27:672–9.
- Haas M, Segev DL, Racusen LC, et al. (2008). Arteriosclerosis in kidneys from healthy live donors: comparison of wedge and needle core perioperative biopsies. Arch Pathol Lab Med 132:37–42.
- Khalil H. (2016). Diabetes microvascular complications – a clinical update. Diabetes Metab Syndr. [Epub ahead of print]. doi: 10.1016/j.dsx.2016.12.022.
- Moshiri M, Moallem SA, Attaranzadeh A, et al. (2017). Injury to skeletal muscle of mice following acute and sub-acute pregabalin exposure. Iran J Basic Med Sci 20:256–9.
- Falconi D, Tattoli F, Brunetti C, et al. (2015). [Rhabdomyolysis from gabapentin: a case report]. G Ital Nefrol. 32(2):pii: gin/32.2.37.
- Barmparas G, Lamb AW, Lee D, et al. (2015). Postoperative infection risk after splenectomy: a prospective cohort study. Int J Surg 17:10–4.
- Andres-Hernando A, Altmann C, Ahuja N, et al. (2011). Splenectomy exacerbates lung injury after ischemic acute kidney injury in mice. Am J Physiol Renal Physiol 301:F907–16.
- Jiang H, Meng F, Li W, et al. (2007). Splenectomy ameliorates acute multiple organ damage induced by liver warm ischemia reperfusion in rats. Surgery 141:32–40.
- Ajmo CT, Jr., Vernon DO, Collier L, et al. (2008). The spleen contributes to stroke-induced neurodegeneration. J Neurosci Res 86:2227–34.
- Wystrychowski W, Filipczyk L, Cierpka L, et al. (2014). Splenectomy attenuates the course of kidney ischemia-reperfusion injury in rats. Transplant Proc 46:2558–61.
- Macedo E, Bouchard J, Soroko SH, et al. (2010). Fluid accumulation, recognition and staging of acute kidney injury in critically-ill patients. Crit Care 14:R82.