Risk factors of chronic kidney disease in cisplatin-based hyperthermia intraperitoneal chemotherapy

Abstract

Purpose

Cisplatin is commonly prescribed in hyperthermic intraperitoneal chemotherapy (HIPEC) for peritoneal malignancy. Acute kidney injury (AKI) is regarded as a common complication after HIPEC combined with cytoreductive surgery (CRS). However, post-HIPEC chronic kidney disease (CKD) is scarce and less investigated. This study aims to investigate the incidence of CKD following cisplatin-based HIPEC and to analyse the associated risk factors.

Materials and Methods

From January 2016 to August 2021, a total of 55 patients treated with CRS and cisplatin-based HIPEC for peritoneal carcinomatosis were categorized retrospectively into groups, with and without CKD. Demographics, comorbidity, surgery, postoperative management, and complications were collected to evaluate risk factors for cisplatin-based HIPEC-related CKD. Univariate and multivariate analyses were conducted to confirm the correlation between different variables and CKD occurrence.

Results

Of the 55 patients, 24 (43.6%) patients developed AKI and 17 (70.8%) patients of these AKI patients progressed to CKD. Multivariate regression analysis identified intraoperative use of parecoxib (Odds Ratio (OR) = 4.39) and intraoperative maximum temperature > 38.5°C (OR = 6.40) as major risk factors for cisplatin-based HIPEC-related CKD occurrence. Though type II diabetes mellitus and intraoperative complications were the independent risk factors of AKI following cisplatin-based HIPEC, but they were not shown in CKD analysis.

Conclusion

Intraoperative use of parecoxib during cisplatin-based HIPEC emerged as a significant risk factor for postoperative CKD. Clinicians should exercise caution in prescribing parecoxib during HIPEC procedures. Additionally, maintaining intraoperative body temperature below 38.5°C might be crucial to mitigate the risk of CKD development. This study underscores the importance of identifying and preventing specific risk factors to improve long-term renal outcomes in patients undergoing cisplatin-based HIPEC.

Keywords:

• Hyperthermic intraperitoneal chemotherapy
• chronic kidney disease
• acute kidney injury
• parecoxib
• cisplatin

Introduction

Acute kidney injury (AKI) had been regarded as unsolved postoperative complication [1]. In patients receiving non-cardiac surgery with low-grade American Society of Anesthesiologist (ASA) physical status, 6% of patients could develop post-operative AKI [2]. AKI has long been thought of a reversible syndrome. But it did increase risk of chronic kidney disease (CKD) development, which followed end-stage renal disease (ESRD), and mortality [3]. However, current studies about postoperative CKD progression remain poor.

Hyperthermic intraperitoneal chemotherapy (HIPEC) following cytoreductive surgery (CRS) treats peritoneal carcinomatosis in various kinds of cancer. It was first used for pseudomyxoma peritonei (PMP) management in 1980 [4]. Compared to systemic chemotherapy, HIPEC owns the advantage of direct intraperitoneal delivery of chemotherapy and enhanced cytotoxic effect to peritoneal surface malignancy [5]. On the contrary, CRS with HIPEC is related to certain complications such as AKI, intra-abdominal abscess, pulmonary complications, anastomotic leakage, intestinal fistula, and hematological toxicity [6]. Rates of major morbidity range from 22-34% and mortality from 0.8-4.1% after CRS and HIPEC [7–9].

Considering hyperthermia-induced splanchnic vasodilation, arterial hypotension, fluid distribution and the direct effects of nephrotoxic chemotherapeutic drugs, HIPEC was recognized as a high-risk factor to postoperative AKI 10. AKI after HIPEC accounts from 1% to 48% and is\ associated with major complications up to 50% [10,11]. Cisplatin is a well-known nephrotoxicity chemotherapeutic drug and has been widely applied as the HIPEC regimen for various cancers. Although many literatures reported cisplatin-based HIPEC as the independent risk factor of postoperative AKI, with incidence of 26.1–48% [10–12], the incidence of post-operative CKD is rarely mentioned.

Parecoxib, a parenteral-specific cyclooxygenase-2 (COX-2) inhibitor, is effective in treating postoperative pain through inhibiting prostaglandin and thromboxane production [13]. COX-2 primarily limit in site of inflammation, while COX-1 mainly found in most body tissue [14]. Therefore, Parecoxib results minor effect on renal function and has been wildly used during surgery to reduce opioid dose by anesthesiologist. Several clinical trials also emphasized the safety of Parecoxib, even in patients with eGFR <90 ml/min·1.73/m² [15,16]. However, the combination use of cisplatin and parecoxib had never been documented.

Therefore, the purpose of this study was to investigate the incidence and predicted risk factors of CKD following cisplatin-based HIPEC and then evaluate the correlation between parecoxib and HIPEC related renal impairment.

Methods

Study population

This retrospective study was approved by the local ethics committee of the China Medical University Hospital with the assigned number of CMUH110-REC2-033. Between Jan 2016 and Aug 2021, we included patients who older than 18 years old underwent CRS and HIPEC for curative intent to treat peritoneal carcinomatosis at our institution. Patients with baseline eGFR less than 60 ml/min/1.73m², extreme dose of cisplatin or undergoing renal replacement therapy were excluded (Figure 1). The recruited patients were divided into two groups, with or without CKD development after HIPEC, for further analysis and then were evaluated the independent risk factors for CKD, which was our primary observed outcome in this study.

Figure 1. Patient allocation flow diagram. Fifty-five patients completed the study. One patient was excluded for extreme dose of cisplatin due to primary tumor of mesothelioma, and another 2 patients were excluded for advanced CKD stage.

We collected data from our prospectively registered database, including patient’s demographic, tumor-related, intraoperative information, and post-operative variables. Nephrotoxicity related factors, such as history of hypertension, diabetes mellitus (DM), hyperlipidemia or use of NSAIDs, ACE inhibitor and contrast media, were also recorded.

Surgical procedure and perioperative care

All patients received CRS procedures for macroscopically detectable peritoneal tumors with complete cytoreduction (CC) score 0 or 1. We also adapted and published laparoscopic CRS for selected cases [17,18]. At the end of surgery, HIPEC (open or close methods) was started by double inflows and outflows drains after intraperitoneal perfused saline (2 L/m²) reached a temperature of 41 to 44 Celsius degrees. The Cisplatin dose and combined regimen were determined by primary neoplasm [19]. In patients with colorectal cancer (CRC) or appendiceal cancer, we used a HIPEC combination of Cisplatin (25 mg/m²/L) plus MMC (3.3 mg/m²/L) for 60–90 min [20]. For ovarian cancer or primary peritoneal serous carcinoma (PPSC) or Mullerian cancer, single Cisplatin (100 mg/m²) [21], or combination of Cisplatin (100 mg/m²) and PTX (175 mg/m²) [22] were administrated for 60 min. We used Cisplatin (75 mg/m²) for 60 min in endometrial cancer [23]. Intraoperative intravenous parecoxib was given according to anesthesiologist’s clinical judgment. During HIPEC, intravenous fluid resuscitation was done to maintain urine output over 1.5 cc/kg/hr. After operation, patients were sent to intensive care unit (ICU) for continuous advanced hemodynamics monitoring. We performed early post-operative extubation and early nutritional support. Postoperative analgesia consisted of patient-controlled epidural or intravenous analgesia and prophylactic antibiotics with cefmetazole was routinely given.

Definition of renal impairment

Renal impairment was analyzed by blood parameters adapting Kidney Disease Improving Global Outcomes (KDIGO) guideline [24]. Acute kidney injury is defined as increase serum creatinine by ≥0.3 mg/dl (≥26.5 μmol/l) within 48 h, or higher than 1.5 times baseline levels in prior 7 days, or urine volume <0.5 ml/kg/h for 6 h; chronic kidney disease is defined as persistent abnormality of kidney function for a period of over 90 days. Creatinine and urea were evaluated on postoperative days (POD) 0,1,2,4,7,30,60,90, and every 6 months till last outpatient department visit. Once if serum creatinine decreased to within the baseline level +0.3 mg/dl or is less than 1.5 times baseline levels, it would not be considered as chronic kidney disease. And postoperative urine output and fluid balance were recorded at least for 5 days. No renal protective medication was regularly administered in patients.

Statistical analysis

Analyses were calculated using SPSS software (IBM SPSS Statistics for Windows, version 26.0, IBM Corp.). Continuous data were reported as the mean ± standard deviation or medians with interquartile ranges and the two groups were compared with Mann-Whitney U test for independent values, respectively. Categorical data were presented as percentages and compared with χ2 test or Fisher’s exact test between groups, as appropriate. Binary logistic regression analysis in a univariable setting was performed to screen possible predictors at first. Parameters with significant correlations were evaluated with multivariate analysis to determinate risk factors of renal impairment. For all results, p-value <0.05 was considered statistically significant.

Results

During the period, 55 patients were recruited with a mean age of 55.4 years old Asians, female-dominated (67.3%), and BMI of 23.84 kg/m² (Table 1). The primary neoplasm type was shown in Figure 2, including 28 colon cancer (51.0%), 12 ovarian cancer (21.8%), 3 endometrial cancer (5.5%), 3 primary peritoneal serous carcinoma (5.5%), 2 appendiceal cancer (3.6%), 2 Mullerian cell carcinoma (3.6%) and 5 others (9%). The overall comorbidities were listed in Table 1.

Figure 2. Distribution of primary neoplasms liocation in 55 patients received cisplatin-based HIPEC.

Table 1. Characteristics of patients with cisplatin-based HIPEC.

Characteristic	No.(%)
Gender
Male	18 (32.7)
Female	37 (67.3)
Age (years)	55.4 ± 11.3
Race (Asian)	55 (100)
BMI	23.8 ± 4.1
Baseline renal function (eGFR)	90.5 ± 20.4
Co-morbidity
Hypertension	12 (21.8)
Diabetes mellitus	7 (12.7)
Hyperlipidemia	3 (5.5)
Cardiopathy	2 (3.6)
COPD	1 (1.8)
Cirrhosis	3 (5.5)
Drugs and other agents during hospital course
Metformin	1 (1.8)
ACE inhibitor & ARB	2 (3.6)
Diuretics	24 (43.6)
NSAID	31 (56.4)
Aminoglycosides antibiotics	0 (0)

In the perioperative hospital course, four patients (7.3%) received iodinated radiological contrast for complete tumor staging or clinical evaluation. The concomitant drugs were prescribed postoperatively with metformin in one patient (1.8%), angiotensin converting enzyme inhibitor in 2 (3.6%), diuretic agents in 24 (43.6%) and NSAID in 31 (56.4%). None of patients received aminoglycosides antibiotics.

Renal impairment

Among all the patients, baseline mean eGFR was 90.5 ± 20.4 ml/min/1.73m² before surgery. There was no patient suffered renal impairment before cisplatin-based HIPEC. Mean eGFR was reduced to 63.9 ± 37.7 ml/min/1.73m² in the short-term after HIPEC therapy and recovered to a mean of 85.8 ± 38.6 ml/min/1.73m² in the long-term period. One patient needed temporary hemodialysis for renal insufficiency. Of the 55 patients, most patients experienced ameliorated eGFR in POD1 and deteriorated in POD2. 24 (43.6%) patients developed AKI in first week and 7 patients recovered in POD7. Rest of 17 (70.8%) patients’ renal function got worsen in 1–2 week after surgery and stabilized in 1 month. Once they meet criteria for CKD in 3 months, no patients recovered in the following days (Figure S1).

Operative features

The operative information of all patients was described in Table 2. and was classified into 2 subgroups based on CKD development. The average dose of Cisplatin was 94.45 ± 14.46 mg/m². The median Peritoneal Cancer Index score (PCI) of all patients was 9 (3–15), 76% had achieved CC-0 and 40% were completed by laparoscopic CRS. Most HIPEC (96.4%) were completed by closed method. Compare between CKD and non-CKD group, there was no significant difference in mean Cisplatin dose, median PCI, CC score, duration of surgery, intraoperative blood loss, blood transfusion, intraoperative complications. Analyzed the intraoperative core body temperature during HIPEC, maximum temperature (Tmax) was statistically significantly higher in CKD group than non-CKD group (38.64 ± 0.59 vs. 38.21 ± 0.64, p = 0.027). But temperature change (△T) did not reveal significantly different in both groups (p = 0.356). Analyzed urine output (mL/kg/hr) during HIPEC, CKD group presented a trend of less urine output than non-CKD group (1.33 ± 1.20 vs. 2.19 ± 1.84, p = 0.056). However, urine output during CRS period did not show any trend. Subgroup analysis of intraoperative complications revealed iatrogenic vessel injury (p = 0.036) with statistically significant difference between CKD and non-CKD groups. In addition, there was no statistically significant difference in post-operative ICU stay (1.59 ± 1.54 vs. 1.42 ± 0.89, p = 0.613) and hospital stay (12.41 ± 2.55 vs. 14.32 ± 9.7, p = 0.431).

Table 2. Operative profile.

Surgery average profile	Overall (n = 55)	CKD group (n = 17)	Non-CKD group (n = 38)	p-value
Mean Cisplatin dose	94.45 ± 14.46	92.65 ± 11.87	95.26 ± 15.55	0.304
Median PCI (percentile range)	9 (3-15)	7 (2.5-13)	10.5 (3.25-15.75)	0.374
CC score				0.484
CC score 0 (%)	42 (76.4)	14 (82.4)	28 (73.7)
CC score 1 (%)	13 (23.6)	3 (17.6)	10 (26.3)
Duration of surgery (min)	647.1 ± 168.3	670.24 ± 135.14	634.76 ± 185.49	0.428
IO Blood loss (mL)	481.2 ± 754.8	618.24 ± 881.86	411.45 ± 713.93	0.482
IO Blood transfusion (unit)	1.65 ± 3.627	2.53 ± 4.125	1.26 ± 3.367	0.317
Maximum of IO temperature (°C)	38.4 ± 0.6	38.64 ± 0.59	38.21 ± 0.64	0.027
UO during HIPEC (mL/kg/hr)	1.93 ± 1.70	1.33 ± 1.20	2.19 ± 1.84	0.056
UO during operation (mL/kg/hr)	1.85 ± 1.13	1.56 ± 0.88	1.98 ± 1.21	0.184
Intraoperative complications
Tumor rupture (%)	2 (3.6)	1 (5.9)	1 (2.6)	0.562
Iatrogenic vessel injury (%)	10 (18.2)	6 (35.3)	4 (10.5)	0.036
Diaphragm perforation (%)	2 (3.6)	0 (0.0)	2 (5.3)	–
Bladder perforation (%)	2 (3.6)	1 (5.9)	1 (2.6)	0.562
Postoperative ICU stay	1.47 ± 1.12	1.59 ± 1.54	1.42 ± 0.89	0.613
Postoperative hospital stay	13.73 ± 8.20	12.41 ± 2.55	14.32 ± 9.7	0.431

CKD: chronic kidney disease; PCI: peritoneal cancer index; CC score: completeness of cytoreductive score; IO: intraoperative; UO: urine output; ICU: intensive care unit.

Risk factors of HIPEC related AKI

As shown in Table 3, univariate and multivariate analysis model confirmed 4 independent factors, diabetes mellitus (OR = 18.72, p = 0.030), Tmax ≥ 38.5 C during HIPEC (OR = 11.76, p = 0.004), intraoperative complications (OR= 5.76, p = 0.041) and use of parecoxib (OR = 5.49, p = 0.030) could be predictors of cisplatin-HIPEC related AKI.

Table 3. Parameter associated to acute kidney injury after cisplatin based HIPEC Obesity, body mass index ≥ 27.

Parameter	Patient No. (n = 55)	Acute Kidney Injury (%)	Univariate Analysis			Multivariate Analysis
			Odds Ratio	95% CI	p	Odds Ratio	95% CI	p
Patient Demographics
Age ≥ 60 years	23	47.8	1.34	0.455 to 3.946	0.596
Male	18	44.4	1.05	0.338 to 3.265	0.933
Diabetes mellitus	7	85.7	10.00	1.112 to 89.905	0.040	18.72	1.335 to 262.465	0.030
Hypertension	12	66.7	3.38	0.875 to 13.021	0.077
Obesity	10	60.0	2.25	0.556 to 9.112	0.256
Hypoalbuminemia	7	57.1	1.87	0.376 to 9.275	0.445
Neoadjuvant chemotherapy	22	59.1	2.89	0.946 to 8.821	0.063
Carboplatin	12	58.3	2.141	0.583 to 7.859	0.251
Cisplatin	2	0	–	–	–
Oxaliplatin	8	87.5	12.353	1.399 to 109.07	0.024
Intraoperative
PCI ≥ 10	26	42.3	0.90	0.310 to 2.626	0.851
Laparoscopic HIPEC + CRS	22	31.8	0.44	0.142 to 1.356	0.153
Massive transfusion	6	66.7	2.90	0.484 to 17.377	0.244
Intraoperative Tmax ≥ 38.5°C	25	64.0	4.89	1.549 to 15.435	0.007	11.76	2.244 to 61.471	0.004
Low UO during operation	27	55.6	2.64	0.880 to 7.910	0.083
Low UO output during HIPEC	30	46.7	1.53	0.496 to 4.724	0.459
Sepsis	12	50.0	1.39	0.385 to 5.013	0.616
Complications	16	43.75	4.40	1.261 to 15.347	0.020	5.76	1.075 to 30.871	0.041
Post-operative
Contrast media	4	50.0	1.32	0.172 to 10.105	0.790
ACE inhibitor/ ARB	3	66.7	2.73	0.232 to 32.008	0.425
Statins	3	66.7	2.73	0.232 to 32.008	0.425
NSAID	5	40.0	0.85	0.130 to 5.529	0.864
Parecoxib	22	63.6	4.03	1.284 to 12.619	0.017	5.49	1.178 to 25.531	0.030

Diabetes mellitus was defined as diabetes symptoms plus a random venous plasma glucose concentration ≥ 11.1 mmol/l or a fasting plasma glucose concentration ≥ 7.0 mmol/l (whole blood ≥ 6.1 mmol/l) or two-hour plasma glucose concentration ≥ 11.1 mmol/l two hours after 75 g anhydrous glucose in an oral glucose tolerance test (OGTT). Hypertension was defined as average systolic blood pressure ≥ 140 mmHg, or when the average diastolic blood pressure ≥ 90 mmHg. Hyperlipidemia was defined as patient taking statin. Obesity was defined as BMI ≥ 27 kg/m². Cirrhosis was defined according to Child-Pugh score and CT imaging. Hypoalbuminemia was defined as serum albumin ≤ 3.5 mg/dl. Neoadjuvant chemotherapy was defined as complete neoadjuvant at least 4 times during past 3 months. Massive transfusion was defined as transfusion more than 10 units of blood during operation. Intraoperative maximum temperature was based on finding in Table 2. Oliguria was defined as average urine output less than 1.5 ml/kg/hr. Sepsis was defined as Surviving Sepsis Campaign guideline. Contrast medium defined as patient received contrast medium for complete evaluation within 1 week before surgery. Parecoxib was defined as patient receiving intravenous parecoxib during cisplatin based HIPEC. Massive transfusion, transfusion of ≥10 units of whole blood or red blood cells. Low urine output < 1.0 ml/kg/hr. PCI: peritoneal cancer index; ACE inhibitor: angiotensin converting enzyme inhibitor; ARB: angiotensin receptor blocker.

Risk factors of HIPEC related CKD

As shown in Table 4, univariate analysis for post-HIPEC CKD disclosed to the neoadjuvant chemotherapy (at least 4 cycles), Tmax >38.5 C during HIPEC, and intraoperative use of parecoxib. In subgroup of neoadjuvant chemotherapy, no significant difference in cisplatin, carboplatin nor oxaliplatin was shown. Other demographic and intraoperative/postoperative variable were not statistically different in CKD development. However, the multivariate analysis only confirmed Tmax >38.5 C during HIPEC (OR = 6.40, p = 0.010) and intraoperative use of parecoxib (OR = 4.39, p = 0.037) as the independent risk factors.

Table 4. Parameter associated to chronic kidney disease after cisplatin-based HIPEC.

Parameter	Patient No. (n = 55)	Chronic kidney disease (%)	Univariate Analysis			Multivariate Analysis
			Odds Ratio	95% CI	p	Odds Ratio	95% CI	p
Patient Demographics
Age > 60 years	23	43.5	2.75	0.848 to 8.902	0.092
Male	18	33.3	1.04	0.315 to 3.450	0.947
Diabetes mellitus	7	57.1	3.24	0.640 to 16.382	0.155
Hypertension	12	50.0	2.58	0.695 to 9.605	0.157
Obesity	10	50.0	2.75	0.675 to 11.204	0.158
Hypoalbuminemia	7	42.9	1.65	0.328 to 8.309	0.544
Neoadjuvant chemotherapy	22	50.0	3.71	1.140 to 12.101	0.029	3.00	0.792 to 11.344	0.106
Carboplatin	12	50.0	2.58	0.695 to 9.605	0.157
Cisplatin	2	0	–	–	–
Oxaliplatin	8	62.5	4.36	0.909 to 20.901	0.066
Intra-operative
PCI ≥ 10	26	26.9	0.70	0.220 to 2.225	0.546
Laparoscopic HIPEC + CRS	22	22.7	0.52	0.151 to 1.750	0.287
Massive transfusion	6	66.7	5.00	0.821 to 30.461	0.081
Temperature max > 38.5°C	25	52.0	5.42	1.568 to 18.718	0.008	6.40	1.568 to 26.100	0.010
Low UO during operation	28	25.0	2.06	0.653 to 6.510	0.217
Low UO during HIPEC	22	18.2	3.00	0.812 to 11.081	0.099
Complications	16	18.75	3.33	0.973 to 11.415	0.055
Post-operative
Sepsis	12	41.7	1.65	0.441 to 6.166	0.458
Contrast medium	4	50.0	2.19	0.282 to 16.946	0.454
ACE inhibitor/ ARB	3	33.3	1.13	0.095 to 13.322	0.926
Statins	3	66.7	4.93	0.415 to 58.575	0.206
NSAID	5	40.0	1.42	0.215 to 9.333	0.717
Parecoxib	22	50.0	3.71	1.140 to 12.101	0.029	4.39	1.095 to 17.600	0.037

Parecoxib subgroup analysis

22 (40%) patients used parecoxib intraoperatively. The baseline eGFR average value in patients with and without use of parecoxib was 93.0 ± 18.5 and 89.9 ± 22.3 ml/min/1.73m² separately. The mean follow-up was 25.4 ± 18.0 months. As shown in Figure 3, reduction of the average eGFR in parecoxib group was 39.5 ± 34.5 and 18.7 ± 39.5 ml/min/1.73m² at 1 week and 3 months after cisplatin-based HIPEC, which showed statistically significantly higher than the eGFR reduction in non-parecoxib group (16.3 ± 34.3 and −3.7 ± 31.7 ml/min/1.73m²).

Figure 3. Postoperative eGFR reduction in parecoxib subgroup. Dash line, reduction of eGFR.

Discussion

CRS and HIPEC has emerged as a viable alternative for treating peritoneal surface neoplasms. Despite its efficacy, HIPEC is associated with notable morbidity, particularly AKI. Existing literature indicates that major abdominal surgery without HIPEC carries a 13.4% AKI incidence [25]. However, when HIPEC is introduced post cytoreductive surgery, the prevalence of AKI exhibits considerable variability, ranging from 0.9% to 48% [6,8–12,26–32]. Liesenfeld had underscored a heightened risk of AKI (OR = 16.29) with cisplatin-based HIPEC regimens [9]. Their research revealed 61.2% of AKI incidence with cisplatin, compared to 14.6% with non-cisplatin HIPEC. Other investigations also reported increased AKI rate (26.1-48%) with Cisplatin-based HIPEC [11,26,27,31,32]. In our study, the AKI rates after cisplatin-based HIPEC was 43.6%, aligning with previous findings. Notably, of the 24 patients with AKI, 70.8% progressed to CKD, a less documented outcome. Our study identified contributors to this high CKD rate, including uncontrolled high body temperature (>38.5 °C) during HIPEC (OR = 6.40) and intraoperative use of parecoxib (OR = 4.39). To our knowledge, relevant data on CKD following HIPEC therapy are lacking.

Cisplatin-based HIPEC, wildly used for peritoneal neoplasms across malignancies, carries a known risk of kidney injury [10,27,31,32]. The complex pathophysiology of cisplatin-induced AKI involves cellular uptake and efflux, apoptosis, vascular injury, oxidative stress, and inflammation [33]. Systemic cisplatin administration results in an overall nephrotoxicity incidence of 23.7%, with 8.1% after the first dose [34]. While intraperitoneal cisplatin is believed to have limited systemic diffusion [35], several literatures associate cisplatin-based HIPEC with AKI and renal impairments are notably linked to cisplatin use [10,26,27,31]. However, conflicting results exist, such as a study by Heakeam reporting only a 3.7% AKI incidence following cisplatin-based HIPEC [30]. Ceresoli et al. demonstrated the HIPEC-related AKI is associated to volume changes during cytoreduction and hyperthermia-induced vasodilatation than cisplatin [28]. Clinical evidence of progressing to CKD after HIPEC is scarce.

In our efforts to mitigate renal impairment, we implemented a program based on Raspe’s and Cooksley’s study [36,37], including pre-operative nutrition status evaluation, perioperative adequate hydration with strict glycemic control, HIPEC perfusate with normal saline-based solution, closely intraoperative urine output monitoring and routine postoperative fresh frozen plasma (FFP) supplement. Sodium thiosulfate was used for nephroprotection in HPIEC, though not introduced in Taiwan. We also avoid the use of nephrotoxic antibiotics and radiocontrast agents in program.

Despite these efforts, some patients still experienced AKI following CRS-HIPEC, with portion progressing to irreversible damage. In the literatures, age, preexisting renal dysfunction, high body mass index, preoperative hypoalbuminemia, hyperglycemia, hypertension, neoadjuvant chemotherapy, and use of NSAID were relatively high risks of post-operative AKI [1,2,10–12,25–27,32,38]. In our study, diabetes mellitus emerged as a significant comorbidity associated with AKI. Although it seems reasonable that the diabetic patients increase risk of renal dysfunction by arterial thickening and glomerulopathic changes, no relevant data had been published. Systemic administration of cisplatin was known as risk factor of nephrotoxicity; therefore, cisplatin was avoided in patients with renal impairment. In subgroup analysis of neoadjuvant chemotherapy regimen, it did not displayed relationship between neoadjuvant chemotherapy and AKI or CKD. Besides, patients receiving cisplatin chemotherapy are mostly afflicted with ovarian cancer, totaling around 20 individuals. Patients receiving cisplatin exhibited normal renal function, and under rigorous control, no association was found between adjuvant cisplatin therapy and CKD. High cisplatin dose, high PCI score, increased blood loss, low intraoperative and postoperative diuresis were intraoperative factors linked to AKI [27,30].

However, only iatrogenic vascular injury during operation increased risk of CKD. Urine output is a sensitive and early marker of kidney perfusion and influenced by several factors, such as hemodynamic, sympathetic tone, aldosterone and antidiuretic hormone concentration during CRS-HIPEC. Some studies demonstrated the intraoperative low urine output was significantly relative to AKI [39,40]. Instead of AKI, our study showed low urine output during HIPEC had a trend to CKD (p = 0.055).

Our study identified 30.9% of patients with cisplatin-based HIPEC developed CKD, with parecoxib use being a significant risk factor for AKI and CKD in Table 4. In the univariate analysis, we overviewed all the medications during admission, especially in the HIPEC course. There was no overall difference in use of metformin, angiotensin converting enzyme inhibitor, angiotensin II receptor blockers, loop diuretics, nor NSAID. Surprisingly, in 22 patients prescribed parecoxib, 14 patients (63.6%) developed AKI (OR = 5.49, p = 0.003). Later, 11 patients of them (50.0%) progressed to CKD (OR = 4.39, p = 0.037). Meanwhile, in patients without parecoxib, 10 patients (30.3%) developed AKI and 6 patients (18.2%) progressed to CKD. 3-month GFR reduction in both groups was 39.5 and 17.6 respectively. Both NSAIDs and COX-2 inhibitors have the potential to elicit renal injury, but most clinical studies disclosed no significant renal injury after COX-2 inhibitors [16,41].

Parecoxib is approved in over 80 countries as perioperative analgesia to prevent or reduce postoperative pain. In our practice, parecoxib was administered intravenously 40 mg per 12 h only during the operation and was not prescribed in ICU or in the general ward. In a pooled analysis study from 9287 patients in 28 comparative trials, the author demonstrated the rate of renal failure was not different between parecoxib (1.0%) and placebo (0.9%) [15]. Currently, no drug-drug interaction between cisplatin and parecoxib has been proven. It is speculated that COX-2 accumulates in kidney after CRS to recovery renal function, but administration of parecoxib hampers the program. The following Cisplatin-based HIPEC deteriorates the situation by renal hypo-perfusion secondary to the hyperthermia-induced vasodilatation and Cisplatin absorption into plasma.

Notable, we found intraoperative maximum core body temperature more than 38.5°C was one of independent risk factors to AKI and CKD. Hendrix et al. also showed the strong correlation (OR = 3.77) between body temperature ≥ 39.5°C and postoperative complications in patients underwent CRS/HIPEC [42]. Ceresoli et al. demonstrated the hyperthermia induced musculocutaneous vasodilatation, stimulated the splanchnic nerve and released massive cytokine into plasma. These situations implicate severe depletion of intravascular volume and result organ hypoperfusion during HIPEC 28. In our study, a total of 25 patients with Tmax ≥ 38.5 °C during HIPEC, 64% had AKI (OR = 11.76) and 52% developed CKD (OR = 6.40). Active cooling of patients has been recommended to prevent intraoperative high body temperature during CRS/HIPEC. However, a significant proportion of patients still develop episodes of intraoperative hyperthermia. In Carlos’ publishment [43], he pointed out longer HIPEC perfusion time and blood transfusions were associated with hyperthermia ≥39 °C.

The aim of this study disclosed the risk factor of nephrotoxicity post cisplatin-based HIPEC therapy. Our data suggested that prescribing parecoxib during cisplatin-based HIPEC therapy causes irreversible renal damage. Prohibit of simultaneous use of parecoxib and cisplatin-based HIPEC should be considered. And presence of intraoperative hyperthermia ≥38.5 °C might be a predictor of post-HIPEC renal dysfunction, which could help to improve patients’ selection to prevent kidney injury.

The limitations conditioned by the retrospective nature of the study can be pointed out [44]. First, inherent bias of patient selection could not be avoided. Second, this was single-center research, and the results need to be verified in multicenter trials. Third, the conclusion from our database is predominately suitable for Asian population. Furthermore, a limited sample size of 55 with low event rate and wide confidence interval was documented.

Data available statement

The data set for this work will be made available on EASY by Data Archiving and Networked Services (DANS): https://doi.org/10.17026/dans-2c7-5v8y

Acknowledgements

We are grateful to Sheng-Chi Chang for leading the program and Hung-Chieh Yeh for alerting increasing CKD prevalence after HIPEC therapy.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors have no funding or financial support to report.