The incidence of cisplatin nephrotoxicity post hyperthermic intraperitoneal chemotherapy (HIPEC) and cytoreductive surgery

Abstract

Background: Cisplatin is commonly used in hyperthermic intraperitoneal chemotherapy (HIPEC) for the management of peritoneal carcinomatosis. Little is known about the nephrotoxic effects of cisplatin use in HIPEC. Objectives: To report the incidence of nephrotoxicity post-HIPEC using cisplatin 50 mg/m² plus doxorubicin 15 mg/m². The incidence of hypomagnesemia was investigated as a secondary endpoint. Methods: This is a retrospective study evaluating patients who received cisplatin with doxorubicin during HIPEC. RIFLE classification was used to assess the development of nephrotoxicity. Variables, such as comorbidities and nephrotoxic medications were obtained. Renal function parameters were also collected, including serum creatinine levels and serum magnesium levels at baseline and at days 3, 7 and 30 after HIPEC. Perioperative urine output (UO) was also recorded. Results: Fifty-three patients were identified. Based on the RIFLE classification, two patients (3.7%) developed acute kidney injury (AKI) following HIPEC with cisplatin. One patient met criteria for renal failure and progressed to chronic renal failure. The other patient had renal injury. Comparable mean creatinine levels were observed at baseline and on day 30 following HIPEC (p > 0.05). The incidence of hypomagnesemia increased to 24.5% by day 7 (p = 0.041) and 30.1% by day 30 (p < 0.001) following HIPEC. Low intraoperative UO, angiotensin II receptor antagonist use and hypertension were associated with development of AKI (p < 0.05). Conclusion: Nephrotoxicity can complicate HIPEC with cisplatin therapy and that permanent renal dysfunction may rarely occur. More attention to be directed toward monitoring magnesium levels after cisplatin use with HIPEC.

• Chemotherapy
• cisplatin
• HIPEC
• nephrotoxicity
• RIFLE

Introduction

Hyperthermic intraperitoneal chemotherapy (HIPEC) following cytoreductive surgery (CRS) has emerged as a new modality for the treatment of patients with peritoneal carcinomatosis (PC) originating from different tumors.1 In this procedure, chemotherapeutic drugs are introduced intraoperatively into the peritoneal cavity at temperature exceeds 39.5 °C.2 The abdominal cavity is perfused for about 60–90 minutes, thereby exposing potential microscopic residual cancer cells directly to the synergistic effects of hyperthermia and cytotoxic agents.1,2 Cisplatin, a platinum-based anticancer agent, exerts its cytotoxic effect by binding and cross-linking DNA. It has been utilized as a cornerstone in many HIPEC protocols at variable doses for the treatment of tumors with PC, including primary peritoneal neoplasms, sarcomas and gynecological tumors.1,3,4 In patients with peritoneal mesothelioma, cisplatin has been independently associated with improved survival versus mitomycin-C during HIPEC.4 Being compatible with other chemotherapeutic agents, cisplatin constitutes a logical component of the multimodality intraperitoneal chemotherapy paradigm. Furthermore, cisplatin penetration and cytotoxicity are augmented by heat, with a thermal enhancement ratio of 2.9 at 41.5 °C.5

A chemotherapeutic regimen that is used at various centers around the world for HIPEC is the combination of cisplatin (50 mg/m²) with doxorubicin (15 mg/m²), infused over 90 minutes.6,7 A recent pharmacokinetic study of this regimen yielded a cisplatin perfusate-to-blood area under the curve ratio of 6.28.8 These data indicate the absorption of cisplatin into the circulation during HIPEC, hence systemic complications cannot be excluded.

Nephrotoxicity is the chief dose-limiting adverse effect of cisplatin treatment. Recent experience shows that systemic cisplatin with saline hydration and diuresis leads to a decrease in creatinine clearance values in more than 18% of subjects.9 Several studies have reported a low incidence of renal function impairment with cisplatin use in HIPEC.1,10,11 However, these studies are inconclusive for several reasons. First, nephrotoxicity definition was not standard among these studies. There was no focus on the description of the acute kidney injury (AKI) pattern and fate, or the development of hypomagnesemia that traditionally complicates cisplatin therapy. Second, these studies used inconsistent cisplatin regimens with various doses, perfusion times and accompanying chemotherapeutic agents. Third, some centers used “early post-surgery chemotherapy” following HIPEC, making it difficult to attribute nephrotoxicity to a single dose of intraoperative cisplatin.12 Finally, factors that independently impair renal function such as an intravenous contrast media, certain nephrotoxic medications and intraoperative renal perfusion were underreported in these studies.

Hypomagnesemia is a common complication of systemic cisplatin therapy, particularly following repeated dosing. It has been observed in over half of cases of all cisplatin-induced nephrotoxicity.13 Cisplatin alters renal tubular handling of magnesium, resulting in significant prolonged dose-related hypomagnesemia, even in the absence of a fall in the glomerular filtration rate (GFR).14

A new concept of AKI definition is RIFLE classification. It provides three grades of severity of renal dysfunction on the basis of a change in serum creatinine levels or the duration and severity of decline in the urine output (UO).15

The aim of this study was to report the incidence of nephrotoxicity in patients who underwent HIPEC using a regimen consisting of cisplatin plus doxorubicin. Pre-exposure and post-exposure assessment of serum magnesium levels was considered a secondary endpoint.

Method

This is a retrospective cohort study of patients who underwent CRS and HIPEC using cisplatin for the treatment of PC from a variety of primary tumors at King Faisal Specialist Hospital and Research Centre, Riyadh. The first patient to receive HIPEC at our tertiary care academic medical center was operated on November 2008. Subsequently, all patients aged 18 years and older who underwent HIPEC using cisplatin from November 2008 to November 2013 were included in this study. At our institution, patients are ineligible for HIPEC with cisplatin if they have a GFR < 45 mL/min/1.73 m², have end-stage renal disease or receive renal replacement therapy.

Data collection

Before starting data collection, the institutional review board approval was obtained. Patient age, sex, weight and height were obtained on admission. Body surface area (BSA) was calculated according to the Mosteller formula. The primary tumor type was also obtained. Potential risk factors for nephrotoxicity were collected; including baseline comorbidities [hypertension, diabetes mellitus and chronic kidney disease (CKD)] and receipt of nephrotoxic agents (e.g. vancomycin, aminoglycosides, amphotericin, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers (ARBs) and nonsteroidal anti-inflammatory drugs) one week prior to one week following HIPEC. All patients were screened for intravenous radiology contrast media recipient within the two weeks prior to surgery or one week after surgery. Serum creatinine, urea, albumin and magnesium levels were obtained during the two days prior to surgery, as well as days 3, 7 and 30 postoperatively. Intraoperative data including surgery duration, amounts of fluid loss and replacement, UO and vasopressors receipt were noted. Postoperative developments of anuria for greater than 8 hours or oliguria were also noted. GFR was calculated using the modified MDRD formula: 186 × (serum creatinine [mg/L^−1.154] × (age [years]) − ^0.203 × (0.742 if female). Patients who developed renal dysfunction as defined by the RIFLE classification (R, I or F; Table 1) were followed for 3 months after cisplatin exposure.

Table 1. RIFLE classification for acute kidney injury.15.

Category	GFR criteria	UO criteria
Risk (R)	Increased creatinine level × 1.5 or GFR decrease > 25%	UO < 0.5 mL/kg/h × 6 h
Injury (I)	Increased creatinine level × 2 or GFR decrease > 50%	UO < 0.5 mL/kg/h × 12 h
Failure (F)	Increased creatinine level × 3 or GFR decrease > 75% or creatinine level > 4 mg/dL	UO < 0.3 mL/kg/h × 24 h or anuria × 12 h
Loss (L)	Persistent acute renal failure or complete loss of renal function for > 4 weeks
ESRD (E)	ESRD for > 3 months

Notes: GFR: glomerular filtration rate; UO: urine output; ESRD: end-stage renal disease.

Chemotherapy

Patients of the study received a standard chemotherapeutic regimen consisting of cisplatin 50 mg/m² (Cisplatin Hospira 100 mg/100 mL ONCO-TAIN®, Hospira, Warwickshire, UK) and doxorubicin 15 mg/m² (Doxorubicin 50 mg per vial, Adriamycin®, Pfizer, Actavis Italy S.p.A 20014 Nerviano, Italy). Ceiling doses of 100 mg for cisplatin and 30 mg for doxorubicin were set. Both chemotherapeutic agents were administered after CRS of the primary tumor. Patients with a GFR of 45–60 mL/min/1.73 m² received 50% reduced dose of cisplatin A carrier solvent of 0.9% NaCl calculated at 2 L/m² was used to fill the abdominal cavity and heated to a temperature of 39.5 °C to 42 °C, measured using four abdominal temperature probes. Cisplatin and doxorubicin were pushed simultaneously into the circulating solvent. The heated solution was circulated for 90 minutes within the open abdomen using the “coliseum technique”. Notably, early postoperative chemotherapy administration post-HIPEC is not practiced at our institution.

Hydration

Before HIPEC, all patients received at least 24 hours of isotonic crystalloid intravenous infusion at 1.5 mL/kg/h. All patients received a 500 mL loading dose of albumin 5% concurrent with isotonic crystalloid solution at anesthesia induction. During the heated perfusion, intravenous fluid administration was increased to maintain a UO of 1 mL/kg/15 minutes. Low-dose dopamine and/or furosemide were administered when there was a drop in UO regardless of the incremental elevation of intravenous fluid infusion rate. Postoperatively, patients were admitted to the intensive care unit (ICU) and immediately initiated on 100 mL of albumin 20%, administrated twice daily for three days, concurrently with at least 3 L of crystalloid solution per day, depending on patient weight.

Nephrotoxicity and hypomagnesemia

The international Acute Dialysis Quality Initiative group suggests the use of RIFLE classification for the definition of AKI (Table 1).15 Oliguria is defined as a UO less than 400 mL per 24 hours, whereas anuria is defined as a UO less than 50 mL per day.16 Hypomagnesemia is defined as serum magnesium levels less than 0.7 mmol/L.

Statistical analysis

Dichotomous data were expressed as frequency distributions and were compared using the χ² or Fisher's exact test. McNemar's test was used to compare dichotomous variables at baseline and at day 30. Normally distributed continuous data were expressed as mean ± SD and were compared using the Student's t-test. Non-normally distributed continuous data were compared using the Mann–Whitney U test. All tests were two-tailed, and a p < 0.05 was considered statistically significant. All analyses were conducted using SAS version 9.3 (Statistical Analysis System, SAS Institute Inc., Cary, NC).

Results

Fifty-three consecutive patients were included in the analysis. Demographics and clinical characteristics are presented in Table 2. The primary tumors included 40 ovarian (75.4%), 6 endometrial (11.3%), 2 liposarcoma (3.7%), 2 mesothelioma (3.7%) and 2 gastrointestinal stromal tumors (3.7%) and 1 cervical carcinoma (1.8%).

Table 2. Demographic data and clinical characteristics for patients with NRIF (no renal failure, injury or risk) versus RIF (renal failure, injury or risk) according to RIFLE classification.

	All patients n = 53	Patients with RIF n = 2	Patients with NRIF n = 51	p Value
Female, %	50 (94.3%)	2 (100%)	48 (94.1%)
Age (years)	49.6 ± 11.1	49.79 ± 11	46 ± 8.4	0.64
Weight (kg)	74.9 ± 14.7	98.1 ± 29	70.3 ± 19	0.01
BSA (m^2)	1.78 ± 0.19	2.06 ± 0.4	1.77 ± 1.8	0.046
Co morbidities
Diabetes mellitus	14 (26.4%)	1 (50%)	13 (25.4%)	0.42
Hypertension	11 (20.7%)	2 (100%)	9 (17.6%)	0.049
CKD	2 (3.7%)	0 (0%)	2 (3.9%)	0.92
Medications
ACE-I	9 (16.9 %)	0(0%)	9 (17.6%)	0.36
ARB	3 (5.6%)	2 (100%)	2 (3.9%)	0.004
NSAIDs	8 (15%)	0 (0%)	8 (15.6%)	0
Vancomycin	25 (47.1 %)	1 (50%)	24 (47%)	0.72
Aminoglycosides	0 (0%)	0%	0%	0
Contrast media	20 (37.7%)	1 (50%)	19 (37.2%)	0.78
Laboratory data
CFR (mL/min/1.73 m^2)	78 ± 1	80 ± 8	77 ± 11	0.1
Intraoperative data
Operating time (h)	11 ± 1.9	10	10.75	0.89
Fluid Balance (L)	9.2 ± 3.9	8.8	9.6	0.09
Albumin 5% (L)	1.9 ± 0.9	2	1.9	0.1
Urine output (mL/h)	358 ± 195	225	364	< 0.02
Vasopressor (n, %)	25 (47.1%)	1	24	0.87

Notes: BSA: body surface area; CKD: chronic kidney diseases; ACE-I: angiotensin converting enzyme inhibitors; ARB: angiotensin II receptor blockers; NSAIDs: non-steroidal anti-inflammatory drugs; and GFR: glomerular filtration rate calculated using MDRD formula.

Based on RIFLE classification, two patients (3.7%) with ovarian cancer developed AKI during their first week following HIPEC with cisplatin, whilst in the ICU. One of them developed renal failure and the other developed renal injury (Table 3 for patient A and B characteristics). There were no statistically significant differences in baseline blood chemistries (creatinine, urea, albumin and magnesium) between patients who developed nephrotoxicity and the rest of the cohort. Hypomagnesemia and hypoalbuminemia were evident on days 3, 7 and 30 post-HIPEC using cisplatin for both patients who developed AKI. However, the mean serum magnesium and albumin levels were comparable between those two patients who that developed AKI and all study patients at all study points (p > 0.05).

Neither two patients with AKI post HIPEC developed anuria or required hemodialysis. Patients A's UO rate post HIPEC was maintained above 0.5 mL/kg/h using loop diuretics administered for pulmonary congestion. Compared to baseline, patient A had elevated creatinine levels (133 mmol/L at one month and 127 mmol/L at three months of HIPEC using cisplatin), and was thus labeled as a CKD patient. Patient B's UO rate decreased to less than 5 mL/kg/h for less than < 12 h on postoperative day 4, whereas her GFR decreased by less than 25% of baseline for three consecutive days. This patient's GFR remained normal at one and three months after HIPEC using cisplatin.

Two patients with CKD before HIPEC were (baseline GFR > 45 mL/min/1.73 m²). Both patients were diabetic and had received reduced cisplatin doses (25 mg/m²). Postoperatively, their creatinine levels did not meet criteria for RIFLE classification for AKI. Four patients in the study had a BSA > 2 m², and hence received the ceiling dose of cisplatin of 100 mg. One of these overweight patients, Patient A, developed nephrotoxicity. Changes in the blood chemistries from baseline versus days 3, 7 and 30 days post-HIPEC using cisplatin for the whole cohort are presented in Table 4.

Table 3. Outcomes of two patients with RIF (renal failure, injury or risk).

Characteristics	Patient A	Patient B	Compared to NRIF P
Nephrotoxicity
RIFLE classification	F	I	Not applicable
Onset (number of days post cisplatin)	6	4	Not applicable
Peak SrCr (mmol/L)	382	101	Not applicable
Time to SrCr peak (days)^a	9	5	Not Applicable
Recovery from AKI (days)^a	14	8	Not applicable
Normalization of SrCr (days)^a	Never	8	Not applicable
Nephrotoxic medications	Vancomycin^b Losartan	Losartan	0.004 for losartan
Intravenous contrast media	Yes	No	NS
Baseline serum magnesium level (mmol/L)	0.73	0.66	NS
Intraoperative data
Urine output (mL/h)	250	200	< 0.025
Fluid balance (L)	11.1	6	NS
Blood loss (L)	0.8	1.8	NS
Vasopressors	no	yes	NS
Albumin (L)	2.5	1.5	NS

Notes: SrCr: serum creatinine; AKI: acute kidney injury; NRIF: no renal failure, injury or risk, based on RIFLE classification; and NS: Non significant.

^aPost cisplatin use with HIPEC.

^bVancomycin trough levels were obtained twice prior to kidney insult 6.1 mg/L on day 2 of vancomycin treatment and 10.5 mg/L on day 4 of vancomycin treatment.

Table 4. Mean serum biochemistries at baseline with comparison at day 3, day 7 and day 30 post-HIPEC using cisplatin.

	Creatinine (mean ± SD) mmol/L	Urea (mean ± SD) mmol/L	Magnesium (mean ± SD) mmol/L	Albumin (mean ± SD) mg/L
Baseline	59.9 ± 13.3	3.7 ± 0.9	0.79 ± 0.07	39 ± 2.8
Day 3 post	43.5 ± 12^a	3.4 ± 1.3^a	0.74 ± 0.14^a	32.3 ± 5.7^a
HIPEC	p < 0.001	p < 0.05	p < 0.01	p < 0.01
Day 7 Post	49.7 ± 16.5^b	5.06 ± 2.5^b	0.77 ± 0.11^b	34.3 ± 4.2^b
HIPEC	p = 0.004	p < 0.01	p < 0.02	p < 0.01
Day 30 Post	58.3 ± 21.1^c	3.6 ± 1.4^c	0.69 ± 0.09^c	36.4 ± 7.2^c
HIPEC	p = 0.4	p > 0.1	p < 0.001	p < 0.025

Notes: ^ap Value for mean baseline versus day 3 post-HIPEC using student t-test.

^bp Value for mean baseline versus day 7 post-HIPEC using student t-test.

^cp Value for mean baseline versus day 30 post-HIPEC using student t-test.

Before HIPEC and cisplatin administration, serum magnesium levels were abnormally low in five patients (9.4%). The incidence of hypomagnesemia was significantly increased to 24.5% and 30.1% after HIPEC on day 7 (p = 0.041) and day 30 (p < 0.001) after HIPEC, respectively. There was no statistically significant difference in the development of hypomagnesaemia between patients that developed nephrotoxicity and those who did not (p > 0.1) at all the study time points.

Discussion

Cisplatin-induced nephrotoxicity is a complex process involving acute cytotoxicity to tubular epithelium, followed by inflammatory cell infiltration and fibroproliferative changes.13 It has been theorized that cisplatin-induced nephrotoxicity varies in a dose-dependent manner.17 High doses of cisplatin causes tubular epithelium necrosis, while low doses induce remove apoptosis of epithelial cells.

Renal toxicity with cisplatin has been noted in 28–36% of patients treated with a single dose of systemic cisplatin at 50 mg/m².¹⁸⁸ Typically, AKI begins several days after cisplatin administration and characterized by elevated serum creatinine and blood urea levels, in the presence of preserved UO. Some patients who develop AKI with cisplatin could progress to CKD.18 Another commonly observed manifestation of cisplatin renal toxicity is hypomagnesemia, which can develop even in the absence of a fall in GFR.19 The risk of renal impairment increases with increasing free serum platinum levels observed with high doses of cisplatin and hypoalbuminemia.20 Preexisting kidney damage, hypomagnesemia and concomitant use of other nephrotoxic agents may potentiate the development of renal impairment.21 Aminoglycosides, amphotericin B, mitomycin-C and non-steroidal anti-inflammatory drugs are known nephrotoxins that could synergize cisplatin nephrotoxicity.13,22

Raspagliesi et al. reported a 5% incidence of renal toxicity among 40 patients who underwent HIPEC using cisplatin at variable doses and combinations, with either mitomycin-C or doxorubicin.10 Zanon et al., reported a 6% incidence of nephrotoxicity in 30 women with ovarian cancer using cisplatin with HIPEC circulated for only 60 minutes.11 In that HIPEC study, the cisplatin dose was 100 mg/m² in nine patients and then increased to 150 mg/m² in the remaining 21 patients. In 16 of these patients, cisplatin was combined with systemic infusion of thiosulfate. In addition, Di Giorgio et al. have reported a comparable risk of nephrotoxicity (4%) after using 75 mg/m² of cisplatin solely in HIPEC.1 A prospective database of HIPEC procedures of variable doses and combinations of cisplatin reported an incidence of nephrotoxicity of 5.4%. With a univariate analysis, cisplatin doses > 240 mg were shown to be associated with poor operative outcomes, a dose that is more than 4 times of what has been used in the current study.23

Some HIPEC treatment studies used the National Cancer Institute Common Terminology Criteria for Adverse Events to assess a wide range of complications, including nephrotoxicity.23 Unlike RIFLE criteria, the assessment of renal function using NCICTE criteria is based solely on GFR.

The low serum creatinine, urea and albumin levels observed postoperatively in this study may be attributable to the extensive fluid administration during and shortly after HIPEC. However, 30 days after HIPEC and cisplatin use, when the most patients were discharged from the hospital, creatinine and urea were comparable to their baseline. The significantly low albumin levels observed one month of post-surgery may reflect the malnutrition that accompanies major surgery.24

In our opinion, the statistically significant higher BSA associated with RIF versus NRIF was not due to higher cisplatin doses, as patients with a BSA more than 2 m² received a ceiling dose of 100 mg of cisplatin. Other factors associated with nephrotoxicity in this study were lower intraoperative UO, use of ARBs and hypertension. Animal studies show losartan use does not alter the onset or severity of cisplatin nephrotoxicity; however, chronic angiotensin II receptor blockade has been associated with improved rate of recovery of renal function in cisplatin-treated rats.25 In contrast, a recent study showed that losartan promote cisplatin-induced renal damage in female rats.26 Although this gender-associated cisplatin nephrotoxicity is consistent with our findings, it is premature to make such conclusion in humans. Chronic systemic hypertension requiring ARBs could induce kidneys ageing, with many studies demonstrating an association between hypertension and cisplatin-related renal toxicity.27,28

It has been shown that cisplatin binding to albumin may attenuate its nephrotoxicity.20 Albumin was generously administrated to our patients during and after HIPEC. However, there was no association between the postoperative hypoalbuminemia and the development of nephrotoxicity.

Although a significant number of our patients developed hypomagnesemia post-HIPEC using cisplatin, the incidence did not reach the 50% level reported with systemic cisplatin use.13 More patients developed low magnesium levels one month of cisplatin exposure compared to one week after cisplatin exposure in this study. This finding may be attributable to the surgical resection of parts of the gastrointestinal tract and altered dietary intake after major surgery.29

The incidence of cisplatin induced nephrotoxicity following HIPEC observed in this study was lower than its systemic administration.18 This could be attributed to the local administration of chemotherapy in HIPEC. On the other hand, this low incidence of nephrotoxicity compared to other HIPEC studies maybe explained by the avoidance of cisplatin-mitomycin-C combination and the efforts of maintaining high UO during surgery.

This study is unique in that the all patients received a standard chemotherapeutic regimen. Furthermore, this study considered many factors that may predispose patients to nephrotoxicity. Using RIFLE classification is considered a modern tool to describe AKI, combing GFR and UO in defining renal impairment. Our finding that ARBs contribute to renal injury, suggests more focus to be directed to hypertensive patients receiving these medications prior to HIPEC using cisplatin. Limitations of this study include its retrospective design and lack of a control group. Furthermore, to truly assess the impact of cisplatin in HIPEC on kidney function, it should be the only exposure to the patient. To report an adverse effect of a medication, it usually requires a study with a higher number of subjects. However, due to the type of the diseases treated with HIPEC and surgically fit patients to undergo this procedure made it difficult to conduct such study on large number of subjects.

In conclusion, this cohort study demonstrates that nephrotoxicity can complicate HIPEC with cisplatin therapy and that permanent renal dysfunction may rarely occur. Furthermore, the RIFLE criteria maybe a useful tool to facilitate data comparison in future studies. More attention to be directed toward monitoring magnesium levels after cisplatin use with HIPEC.

Declaration of interest

The authors declare no conflict of interest.