Effect of Urine-guided Intraoperative Hydration on Incidence of Acute Kidney Injury after Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy for Pseudomyxoma Peritonei: A Randomized Trial

Abstract

Background:

Acute kidney injury (AKI) is common after cytoreductive surgery (CRS) and hyperthermia intraperitoneal chemotherapy (HIPEC). Urine-guided hydration is found effective in preventing AKI in other high-risk patient populations. This study tested whether targeted intraoperative urine output maintenance and simultaneous hydration can reduce AKI in patients after CRS-HIPEC.

Methods:

In this randomized trial, adult patients who were scheduled to undergo CRS and cisplatin-based HIPEC for pseudomyxoma peritonei were randomized to receive either urine-guided hydration (urine output greater than or equal to 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ or greater than or equal to 200 ml ⋅ h⁻¹) or routine hydration (urine output greater than or equal to 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹) during the procedure. The primary endpoint was the incidence of AKI within 7 days after surgery, diagnosed according to the Kidney Disease Improving Global Outcome criteria.

Results:

From July 24, 2023, to July 18, 2024, 168 patients (mean age, 58 yr; 66.1% female sex) were enrolled and randomized; all were included in the intension-to-treat analysis. AKI incidence within 7 days was lower with urine-guided hydration than with routine hydration (21.4% [18 of 84] vs. 39.3% [33 of 84]; relative risk [RR], 0.55; 95% CI, 0.33 to 0.89; P = 0.012). Subgroup analysis showed that the proportion of AKI diagnosed according to urine criteria (urine output less than 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ for 6 h or longer) was less with urine-guided hydration than with routine hydration (21.4% [18 of 84] vs. 35.7% [30 of 84]; RR, 0.60; 95% CI, 0.36 to 0.99; P = 0.040). Patients with urine-guided hydration developed fewer major complications within 30 days (36.9% [31 of 84] vs. 56.0% [47 of 84]; RR, 0.66; 95% CI, 0.47 to 0.92; P = 0.013). Adverse events did not differ between groups.

Conclusions:

In patients who underwent CRS and cisplatin-HIPEC for pseudomyxoma peritonei, intraoperative urine-guided hydration reduced postoperative AKI by more than 40% and was safe. A large trial is warranted to verify the results of this study.

Randomizing patients with pseudomyxoma peritonei who underwent hyperthermia intraperitoneal chemotherapy to a urine output–guided hydration strategy yielded an absolute risk reduction for acute kidney injury within 7 days after surgery of about 18% (number needed to treat = 6), compared with routine care. Patients in the urine-guided group received about 1 l more crystalloid solution and more than 90% received furosemide (compared with 50% in the routine care group), leading to a median 1-l greater urine output during surgery. Randomization to urine-guided care also was associated with fewer composite complications within 30 days of surgery (number needed to treat = 5), particularly driven by reduced acute kidney injury and new-onset arrhythmia.

Editor’s Perspective

What We Already Know about This Topic

• Acute kidney injury (AKI) after surgery is associated with a wide range of adverse outcomes, but interventions to reduce AKI have not been universally effective

• Hyperthermic intraperitoneal chemotherapy is an emerging strategy to extend life in patients with certain peritoneal malignancies but has a particularly high rate of AKI

What This Article Tells Us That Is New

• Randomizing patients with pseudomyxoma peritonei who underwent hyperthermia intraperitoneal chemotherapy to a urine output–guided hydration strategy yielded an absolute risk reduction for AKI within 7 days after surgery of about 18% (number needed to treat = 6), compared with routine care

• Patients in the urine-guided group received about 1 l more crystalloid solution and more than 90% received furosemide (compared with 50% in the routine care group), leading to a median 1-l greater urine output during surgery

• Randomization to urine-guided care also was associated with fewer composite complications within 30 days of surgery (number needed to treat = 5), particularly driven by reduced AKI and new-onset arrhythmia

Pseudomyxoma peritonei is a rare peritoneal malignancy characterized by progressive mucinous ascites and diffused implant.^1–3 The recommended management is a combination of cytoreductive surgery (CRS) and hyperthermia intraperitoneal chemotherapy (HIPEC)^4,5 that is surgical resection of macroscopic tumors by multivisceral resections and/or extensive peritonectomy followed with eradication of residual microscopic tumors or free cancer cells by circulating heated chemotherapy drugs in the abdominal cavity.⁶ This integrated surgical–medical strategy changes management for pseudomyxoma peritonei from a palliative care to a potentially curative therapy.^4,5 However, CRS-HIPEC is generally associated with increased complications.⁷ Indeed, patients treated with cisplatin-based HIPEC have a high incidence of acute kidney injury (AKI), ranging from 3.7 to 48.4%; those given cisplatin-based HIPEC were at especially higher risk.^8–10

It is confirmed that even a mild increase in creatinine levels is associated with adverse outcomes.¹¹ Measures that prevent AKI occurrence may improve outcomes in this patient population. Among available studies, Hakeam et al.⁸ reported the lowest incidence of 3.7% in 53 patients who were diagnosed with peritoneal metastatic cancer and treated with CRS and cisplatin-HIPEC; the low AKI incidence was attributed to the effect of maintaining a higher intraoperative urine output. Among patients who underwent CRS and cisplatin-HIPEC for peritoneal metastasis originating from ovarian cancer, Angeles et al.⁹ also found that higher intraoperative urine output was associated with less AKI; however, they did not explore the cutoff value of intraoperative urine output to predict AKI occurrence.

Urine management is found effective for preventing AKI in some clinical settings. For example, adequate volume repletion combined with diuretics to reach a target urine output of approximately 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ was suggested for prevention and treatment of rhabdomyolysis-induced AKI.¹² Combining high-dose furosemide and matched real-time rehydration with a target urine output at 300 ml ⋅ h⁻¹ or higher was found effective in preventing contrast-induced AKI.^13,14 A strategy of balanced forced diuresis aiming a urine output of at least 200 ml ⋅ h⁻¹ also reduced AKI in high-risk patients after cardiac surgery with cardiopulmonary bypass.¹⁵ For patients undergoing CRS-HIPEC, it is recommended to maintain urine output at 1 ml ⋅ kg⁻¹ ⋅ h⁻¹ during CRS and reconstruction and at 2 ml ⋅ kg⁻¹ ⋅ h⁻¹ during HIPEC¹⁶ or at least 1 ml ⋅ kg⁻¹ ⋅ h⁻¹ during the whole procedure.¹⁷ However, evidence remains limited in this aspect.¹⁸ In this study, we conducted a randomized trial to test the hypothesis that maintaining intraoperative urine output greater than or equal to 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ or greater than or equal to 200 ml ⋅ h⁻¹ with diuretics in combination with simultaneous hydration might reduce AKI in patients undergoing CRS-HIPEC for pseudomyxoma peritonei.

Materials and Methods

Study Design

This was a randomized trial with two parallel arms. The trial was approved by the local ethics committee (approval No. 2023-080; June 19, 2023), registered with the ClinicalTrials.gov (NCT05939193; Dong-Xin Wang; July 2, 2023), and conducted in Aerospace Center Hospital. Written informed consent was obtained from each participant before enrollment. This article adhered to the applicable Consolidated Standards of Reporting Trials guidelines.

Participants

Potential participants were screened the day before surgery. We enrolled adult patients (aged 18 yr or older) who were diagnosed with pseudomyxoma peritonei; experienced an interval of at least 14 days since the last chemotherapy, immunotherapy, or radiotherapy; and were scheduled for CRS-HIPEC. We excluded those who had persistent atrial fibrillation, new-onset cardiovascular event (acute coronary syndrome, stroke, or congestive heart failure) within 3 months, hemodynamic instability requiring vasopressors, a history of hypersensitivity to furosemide, or a diagnosis of end-stage renal disease.

Randomization and Masking

A random list was generated in a 1:1 ratio with permuted-block sizes by an independent statistician using the SAS 9.0 statistical software package. The generated random numbers were sealed in consecutively numbered opaque envelopes, which were preserved by a research assistant (N.L.) who otherwise was not involved in the trial. For a specific participant, the research assistant opened the random envelopes shortly before anesthesia according to the recruitment sequence, and the responsible anesthesiologist(s) performed intraoperative management according to the randomization results. In this way, the enrolled patients were randomized to receive either urine-guide hydration (urine-guided group) or routine hydration (control group). Investigators who were responsible for data collection and postoperative follow-up (J.-K.Y., Y.-Y.X., T.W., and S.-C.G.) were not involved in anesthesia and perioperative care and were not aware of intraoperative management. All patients, healthcare team members aside from responsible anesthesiologists, and investigators for outcome assessment were blinded to study group allocation.

Intervention, Procedures, and Perioperative Care

No preanesthesia medication was given. Renal protectants and preoperative hydration were not provided.¹⁹ Standard intraoperative monitoring included electrocardiography, pulse oximetry, concentrations of inhalational anesthetics and expired carbon dioxide, Bispectral Index, nasopharyngeal temperature, invasive blood pressure and stroke volume variation (Vigileo/FloTrac system, Edwards Lifesciences, USA), and hourly urine output.¹⁶ Arterial blood gas and central venous pressure were measured when necessary.

All patients received general anesthesia with endotracheal intubation. Anesthesia was induced with intravenous sufentanil and/or remifentanil, propofol or ciprofol, and rocuronium or cis-atracurium and was maintained with intravenous infusions of propofol and remifentanil and inhalation of sevoflurane, supplemented with muscle relaxants (cis-atracurium). The target was to maintain a Bispectral Index between 40 and 60. Peripheral nerve block was not routinely performed during the study period.

As a routine practice, intravenous infusion of crystalloid solution (lactated Ringer’s solution and normal saline) was initiated from the arrival in the operating room and continued throughout the procedure. A volume of 250 ml of crystalloids was infused in 10 min in case of a stroke volume variation greater than 10%; the target was to maintain stroke volume variation at or below 10%. Gelatin and/or 20% albumin was infused for volume replacement if required; the total volume of gelatin was restricted to less than or equal to 1,000 ml. Blood products were transfused to maintain hemoglobin concentration higher than 7 g ⋅ dl⁻¹. Vasoactive drugs were administered when necessary; blood pressure was maintained within 30% from baseline.

For patients assigned to receive urine-guided hydration, urine output was maintained at greater than or equal to 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ or greater than or equal to 200 ml ⋅ h⁻¹ during the intraoperative period; intravenous furosemide was administered if required (initial bolus of 10 to 20 mg, followed by continuous infusion at a rate of 10 mg ⋅ h⁻¹, adjusted as appropriate). For patients assigned to receive routine hydration, urine output was maintained according to routine practice, which was greater than or equal to 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ during the intraoperative period; intravenous furosemide was only allowed in case of persistent urine output of less than 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ or at the discretion of the responsible anesthesiologists. For patients of both groups, intraoperative fluid infusion was managed as above, according to stroke volume variation and clinical monitoring.

CRS and cisplatin-based HIPEC were performed according to Chinese expert consensus and local routine.²⁰ CRS as described by Sugarbaker²¹ was first performed according to clinical judgment of the attending surgeons. HIPEC was then performed using a closed technique. Specifically, peritoneal perfusion inflow and outflow catheters were placed before temporary closure of the abdominal cavity. The catheters were connected to a perfusion machine (RHL-2000B; Jilin Maida Technology Development Co., Ltd., China), which heated and circulated the perfusion fluid with an inflow temperature of 43.5°C and an outflow temperature of 41.0°C, at a rate of approximately 800 to 1,000 ml ⋅ min⁻¹, and for a duration of 60 min. The dose of cisplatin was 60 to 80 mg based on body surface area; the volume of perfusate was 3,500 to 4,000 ml. After HIPEC, the abdominal cavity was reopened for digestive tract reconstruction, drainage tube placement, and sufficient hemostasis and then definitively closed.

Patients who were stable during the procedure were extubated and monitored in the postanesthesia care unit and transferred back to the general wards; otherwise, patients were admitted to the intensive care unit (ICU) for further monitoring and treatment. A patient-controlled intravenous analgesia pump was provided for postoperative analgesia. Early postoperative intraperitoneal chemotherapy was typically conducted from 2 to 6 days as appropriated. Other perioperative managements including fluid infusion and diuretic use were performed according to the local routine.

Data Acquisition

Baseline data included demographic characteristics, preoperative comorbidities, baseline blood pressure and heart rate, main laboratory results, tumor characteristics, and history of tumor therapy. General status were evaluated with the New York Heart Association classification, American Society of Anesthesiologists (Schaumburg, Illinois) physical status, Charlson Comorbidity Index,²² and modified frailty index (measured according to 11 items; scores range from 0 to 1, with higher score indicating more severe frailty).²³ Activity of daily living was evaluated with the Barthel Index (scores range from 0 to 100, with higher score indicating better activity).²⁴ Nutritional status was evaluated with the nutritional risk screening 2002 (NRS2002; a five-item scale including age, body mass index, appetite, accidental weight loss, and severity of acute illness; total scores range from 0 to 7, with higher score indicating higher risk of malnutrition).²⁵ The previous surgical score was evaluated (scores range from 0 to 3, where 0 indicates no surgery in abdominal region or only biopsy, 1 indicates exploratory laparotomy in one to two abdominal regions, 2 indicates exploratory laparotomies with resection in two to five abdominal regions, and 3 indicates extensive previous cytoreduction in more than five abdominal regions).²⁶

Intraoperative data were collected. Preoperative cancer distribution was evaluated with the peritoneal cancer index. The index divided abdominopelvic cavity into nine regions and the small bowel into four parts; each region or part was scored from 0 to 3 according to tumor volume, resulting a total score from 0 to 39, with a higher score indicating higher tumor burden.²⁶ Other surgery-related data included volumes of mucus and ascites, duration and type of surgery, and dose of cisplatin during HIPEC. Completeness of cytoreduction was evaluated with a 4-point scale, where 0 indicates no macroscopic residual nodule, 1 indicates residual nodule less than or equal to 2.5 mm, 2 indicates residual nodule greater than 2.5 mm but less than or equal to 2.5 cm, and 3 indicates residual nodule greater than 2.5 cm; a score of 0 or 1 was defined as complete cytoreduction.²⁶ Anesthesia-related data included duration of anesthesia, types and doses of medications, infused/transfused and lost (including estimated blood loss and urine output) volumes, and use of vasoactive drugs. Vital signs including blood pressure, heart rate, and nasopharyngeal temperature were extracted from the anesthesia information system.

For AKI assessment after surgery, patients were visited three times daily (8 to 10 am, 3 to 4 pm, and 6 to 8 pm) during the first 7 days by a specially trained investigator (J.-K.Y.; anesthesia nurse). For patients admitted to the ICU, hourly urine output was documented in the electronic medical record. For patients in general wards, hourly urine output was recorded by bedside caregivers (mainly family members who were blinded to group assignment) until removal of urinary catheter; missing data were replaced with the average values between the nearest timepoints. Average hourly urine output was calculated after urinary catheter removal. Serum creatinine was suggested to be measured daily during the first 7 days but could be measured every other day if the first two test results were normal. We collected data including hourly urine output, volume of daily infused fluid, use of diuretics, and serum creatinine within 7 days, as well as and the highest creatinine before discharge. We also documented the conduct of intraperitoneal chemotherapy during the early postoperative period.

For delirium and other endpoint assessments after surgery, patients were visited twice daily (8 to 10 am and 6 to 8 pm) during the first 4 days, once daily from postoperative days 5 to 7, and on the 30th postoperative day by specifically trained investigators (anesthesiologists Y.-Y.X., T.W., and S.-C.G.). Delirium was assessed using the Confusion Assessment Method for nonintubated patients²⁷ or the Confusion Assessment Method for the Intensive Care Unit for intubated patients²⁸ until the 7th postoperative day. Delirium occurrence was defined as at least one positive episode during this period. We collected biochemical test results within 7 days, including serum levels of high-sensitivity troponin I and bilirubin. Myocardial injury was defined as a peak high-sensitivity troponin I level exceeding the 99th percentile of the upper reference limit.²⁹ Acute hepatic injury was defined as a rise in serum total bilirubin greater than or equal to 33 µmol ⋅ l⁻¹.³⁰ Major complications were generally defined as new-onset conditions that were deemed harmful and required therapeutic intervention within 30 days, i.e., grade 2 or higher on the Clavien–Dindo classification.³¹

Study Endpoints

Our primary endpoint was the incidence of AKI within 7 days after surgery, diagnosed according to the Kidney Disease Improving Global Outcome (KDIGO) criteria.³² Specifically, patients were considered to have AKI if their serum creatinine values increased by either greater than or equal to 26.5 µmol ⋅ l⁻¹ within 48 h or greater than or equal to 1.5 times baseline within 7 days or urine volume less than 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ for 6 h or longer. The severity of AKI was also classified according to the KDIGO criteria: stage 1 refers to increase in creatinine by greater than or equal to 26.5 µmol ⋅ l⁻¹ or 1.5 to 1.9 times baseline or urine volume less than 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ for 6-12 h; stage 2 refers to increase in creatinine 2 to 2.9 times baseline or urine volume less than 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ for 12 h or longer; stage 3 refers to increase in creatinine by greater than or equal to 353.6 µmol ⋅ l⁻¹ or three or more times baseline, initiation of renal replacement therapy, urine volume less than 0.3 ml ⋅ kg⁻¹ ⋅ h⁻¹ for 24 h or longer, or anuria for 12 h or longer.³² The baseline creatinine level was defined as the most recent creatinine value before surgery.

Secondary endpoints included classification of AKI within 7 days, ICU admission, other organ injuries within 7 days (including delirium, myocardial injury, and hepatic injury), length of hospital stay, major complications within 30 days, and all-cause 30-day mortality after surgery. As exploratory endpoints, we reported time to gastric tube removal, time to defecation, and length of complete intravenous nutrition after surgery. We also calculated the comprehensive complication index (scores range from 0 [no postoperative complication] to 100 [death]) for each patient, which summarizes all postoperative complications according to the Clavien–Dindo classification.³³ Long-term outcomes will be reported separately.

Adverse Events

Adverse events were monitored from anesthesia induction until discharge from postanesthesia care unit or 2 h after ICU admission. Specifically, we evaluated tachycardia (heart rate greater than 100 beats ⋅ min⁻¹ or an increase of more than 30% from baseline), bradycardia (heart rate less than 40 beats ⋅ min⁻¹ or a decrease of greater than 30% from baseline), hypotension (systolic blood pressure less than 80 mmHg or a decrease of greater than 30% from baseline), hypertension (systolic blood pressure greater than 180 mmHg or an increase of greater than 30% from baseline), hypothermia (nasopharyngeal temperature less than 36°C), hyperthermia (nasopharyngeal temperature greater than 38°C during CRS or greater than 39°C during HIPEC), new-onset arrhythmia (premature atrial or ventricular contraction greater than 5 beats ⋅ min⁻¹, supraventricular or ventricular tachycardia, and atrial fibrillation), massive hemorrhage (blood loss greater than 1,500 ml or 30% of total blood volume), desaturation (pulse oxygen saturation less than 90% in room air), and postoperative nausea and vomiting (any retching, vomiting, or requirement for antiemetics).

Statistical Analysis

Sample Size Estimation

An observational study in patients with ovarian peritoneal carcinomatosis reported a 48% incidence of AKI after CRS-HIPEC.⁹ In a randomized trial, urine flow rate (greater than or equal to 300 ml ⋅ h⁻¹)–guided hydration reduced contrast-relative AKI by 44%.³⁴ For this trial, we hypothesized that AKI would occur in 48% of our patients and that urine-guided hydration would reduce AKI by 44% (AKI incidence decreased from 48 to 27%). A total of 168 patients (84 in each group) were required to detect the difference with a power of 80% at a significance level of 0.05 and an expected dropout rate of 5%. Sample size was calculated using tests for two proportions PASS version 15 software.

Outcome Analysis

Outcome analyses were primarily performed in an intention-to-treat population, that is, all patients were analyzed in the groups to which they were assigned. For the primary endpoint, analysis was also performed in a per-protocol population, excluding those with major protocol deviations. Missing data were not replaced.

The balance of baseline variables between the two groups was assessed using absolute standardized differences, defined as absolute differences in means, mean ranks, or proportions divided by the pooled SD. Baseline variables with an absolute standardized difference of greater than 0.302 were considered imbalanced using the cutoff point recommended by Austin ($1\mathrm{.96}\times \sqrt{(\mathrm{n}1+\mathrm{n}2)/(\mathrm{n}1\times \mathrm{n}2)}$)³⁵ and adjusted in subsequent analysis when considered necessary. Comparisons of intraoperative data and adverse events between groups were carried out using Student’s t test or Mann–Whitney U test for continuous variables, and either chi-square or Fisher exact test for categorical variables, as appropriate.

For primary endpoint, the incidence of AKI within 7 days was compared with a chi-square test, with difference between groups expressed as relative risk (RR) and 95% CI. A similar analysis was performed for the per-protocol population. Diagnostic bases for AKI were analyzed with continuity-corrected chi-square or Fisher’s exact tests; RR and 95% CI were calculated.

For secondary endpoints and exploratory results, categorical variables (classifications of AKI, incidence of other organ injuries within 7 days, incidence of major complications within 30 days, and 30-day mortality after surgery) were analyzed with chi-square, continuity-corrected chi-square, or Fisher’s exact tests; differences were expressed as RRs and 95% CIs. Continuous variables such as comprehensive complication index was analyzed using asymptotic Wilcoxon rank sum test; median difference and 95% CI were calculated with Hodges–Lehmann estimators. Time-to-event results (duration of mechanical ventilation, lengths of stay in ICU and hospital after surgery, time to gastric tube removal, time to defecation, and length of complete intravenous nutrition after surgery) were analyzed with Kaplan–Meier survival analyses with differences between groups tested using log-rank tests; univariable Cox proportional hazards models were used to calculate hazard ratios and 95% CIs. The proportional hazards assumption was evaluated with the Schoenfeld residual test. Patients who died within 30 days or stayed in the ICU or hospital for more than 30 days were censored at the time of death or 30 days after surgery.

A two-sided P value < 0.05 was considered statistically significant. Statistical analyses were performed with SPSS 27.0 (IBM SPSS, Inc., USA) and SAS 9.4 (SAS Institute Inc., USA).

Results

Patients

From July 24, 2023, to July 18, 2024, 313 patients were screened for eligibility. Of these, 183 patients met the inclusion/exclusion criteria; 168 patients gave consents and were randomized to receive either urine-guided hydration (urine-guided group; n = 84) or routine hydration (control group; n = 84). There were four major protocol deviations during the study period, with three patients in the urine-guided group and one patient in the control group. Therefore, all 168 patients were included in the intention-to-treat analysis; 164 patients were included in the per-protocol analysis, with 81 cases in the urine-guided group and 83 in the control group (fig. 1). Follow-ups for perioperative outcomes ended on August 18, 2024.

Fig. 1.

Flowchart of the trial. CRS, cytoreduction surgery; HIPEC, hyperthermia intraperitoneal chemotherapy.

Of all enrolled patients, mean (± SD) age was 58 (± 10) yr, and 66.1% (111 of 168) were female. Baseline variables were generally balanced between the two groups, except that the mean heart rates were slightly lower, serum sodium concentration was slightly higher, and the proportion with tumor of appendiceal origin were slightly higher in the urine-guided group than in the control group (table 1).

Table 1.

Baseline Data

Characteristic	Urine-guided (n = 84)	Control (n = 84)	Absolute Standardized Difference
Demographic data
Age, mean ± SD, yr	58 ± 9	58 ± 11	0.061
Female sex, No. (%)	53 (63.1)	58 (69.0)	0.123
Body mass index, mean ± SD, kg/m2	23.9 ± 3.4	24.1 ± 3.1	0.043
Comorbidity
Nervous system diseases, No. (%)	3 (3.6)	2 (2.4)	0.063
Previous stroke/transient ischemic attack	1 (1.2)	2 (2.4)	0.109
Others*	2 (2.4)	0 (0.0)	0.155
Cardiovascular system, No. (%)	29 (34.5)	28 (33.3)	0.025
Hypertension	23 (27.4)	22 (26.2)	0.026
Coronary heart disease	3 (3.6)	4 (4.8)	0.063
Arrhythmia	7 (8.3)	4 (4.8)	0.128
Cardiomyopathy	0 (0.0)	1 (1.2)	0.154
Perivascular diseases	2 (2.4)	3 (3.6)	0.077
Respiratory diseases, No. (%)	18 (21.4)	24 (28.6)	0.173
COPD/bronchiectasis	2 (2.4)	1 (1.2)	0.078
Previous pulmonary embolism	0 (0.0)	1 (1.2)	0.155
Pleural effusion	4 (4.8)	5 (6.0)	0.056
Atelectasis	6 (7.1)	8 (9.5)	0.092
Others†	7 (8.3)	9 (10.7)	0.086
Metabolism, No. (%)	7 (8.3)	9 (10.7)	0.086
Diabetes	12 (14.3)	12 (14.3)	<0.001
Hyper-/hypothyroidism	6 (7.1)	2 (2.4)	0.183
Renal diseases, No. (%)	11 (13.1)	11 (13.1)	< 0.001
Chronic kidney dysfunction (stage 4 or lower)	11 (13.1)	11 (13.1)	< 0.001
Others‡	1 (1.2)	2 (2.4)	0.109
Chronic hepatic diseases, No. (%)§	9 (10.7)	9 (10.7)	<0.001
General status
NYHA class, No. (%)			0.150
Ⅰ	68 (81)	63 (75)
II	16 (19)	21 (25)
ASA physical status, No. (%)			0.201
II	66 (78.6)	59 (70.2)
III	18 (21.4)	25 (29.8)
Charlson comorbidity index, median (IQR), points‖	8 (8, 8)	8 (8, 8)	0.033
Modified frailty index, median (IQR), points#	0 (0, 0.09)	0 (0, 0.09)	0.026
Barthel index, median (IQR), points**	100 (100, 100)	100 (100, 100)	0.281
NRS2002, median (IQR), points††	2 (2, 3)	2 (2, 3)	0.167
Parenteral nutrition, median (IQR), days	1 (1, 1)	1 (1, 1)	0.135
Physical examination
Mean arterial pressure, mean ± SD, mmHg	93 ± 8	92 ± 10	0.091
Heart rate, mean ± SD, beats/min	78 ± 7	81 ± 8	0.366
Muscular calf vein thrombosis, No. (%)	7 (8.3)	8 (9.5)	0.043
Latest laboratory tests
Creatinine, median (IQR), µmol/l	67.1 ± 14.6	64.2 ± 19.6	0.210
Urea nitrogen, median (IQR), mmol/l	4.5 (3.7, 5.3)	4.7 (3.6, 5.7)	0.128
eGFR, median (IQR), ml ⋅ min−1 ⋅ 1.73 m−2	96 (83, 103)	97 (89, 104)	0.124
Na+, median (IQR), mmol/l	140 (139, 141)	139 (138, 141)	0.305
Mg2+, median (IQR), mmol/l	0.8 (0.8, 0.9)	0.8 (0.8, 0.9)	0.103
Albumin, mean ± SD, g/l	37 ± 3	37 ± 4	0.233
Hemoglobin, median (IQR), g/l	123 (112, 134)	118 (104, 128)	0.238
Tumor characteristics
Type of histopathology, No. (%)			0.267
Low grade	56 (66.7)	48 (57.1)
High grade	26 (31.0)	36 (42.9)
Others	2 (2.4)	0 (0.0)
Pseudomyxoma peritonei, No. (%)	82 (97.6)	84 (100)	0.155
Appendiceal origin, No. (%)	77 (91.7)	68 (81.0)	0.385
History of tumor therapy
Previous surgical score, No. (%)‡‡			0.239
0	13 (15.5)	16 (19.0)
1	13 (15.5)	10 (11.9)
2	24 (28.6)	17 (20.2)
3	34 (40.5)	41 (48.8)
Systemic chemotherapy, No. (%)	15 (17.9)	22 (26.2)	0.216

Absolute standardized differences in bold indicate values greater than 0.302, which are considered imbalanced between the two groups.

^*Included multiple sclerosis and Guillain–Barré syndrome.

^†Included pulmonary cancer, tuberculosis, interstitial changes, and inflammatory changes.

^‡Included kidney cancer and unilateral renal agenesis.

^§Included hepatitis, cirrhosis, and Child–Pugh class B and C.

^‖Evaluated according to the 1987 version.²²

^#A 11-item scale; scores range from 0 to 1, with higher score indicating more severe frailty.²³.

^**Score ranges from 0 to 100, with a higher score indicating better function.²⁴

^††A five-item scale including age, body mass index, appetite, accidental weight loss, and severity of acute illness. Total score ranges from 0 to 7, with higher score indicating higher risk of malnutrition.²⁵

^‡‡Scores range from 0 to 3; 0 indicated no surgery in abdominal region or only biopsy, 1 indicated exploratory laparotomy in one or two abdominal regions, 2 indicated exploratory laparotomies with resection in two to five abdominal regions, and 3 indicated extensive previous cytoreduction in more than five abdominal regions.²⁶

ASA, American Society of Anesthesiologists; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; IQR, interquartile range; NRS2002, nutritional risk screening 2002; NYHA, New York Heart Association.

During the intraoperative period, patients in the urine-guided group received more crystalloid infusion (median 7,100 ml [interquartile range (IQR), 5,750 to 8,600] vs. median 6,100 ml [IQR, 5,100 to 7,525]; P = 0.008), were given more diuretics (use of furosemide: 91.7% [77 of 84] vs. 51.2% [43 of 84]; P < 0.001; dose of furosemide: median 10.0 mg [IQR, 4.5 to 14.9] vs. median 2.0 mg [IQR, 0.0 to 10.0]; P < 0.001), and had more urine output (median 2,200 ml [IQR, 1,800 to 2,650] vs. median 1,280 ml [IQR, 840 to 1,602]; P < 0.001) than those in the control group. Other intraoperative variables including fluid balance (median 5,700 ml [IQR, 4,370 to 6,920] vs. median 5,825 ml [IQR, 4,662 to 6,980]; P = 0.598) were comparable between the two groups (table 2).

Table 2.

Intra- and Postoperative D7

Data	Urine-guided (n = 84)	Control (n = 84)	P Value
Surgery-related data
Peritoneal cancer index, median (IQR), points*	33 (12, 37)	33 (21, 36)	0.944
Mucus, median (IQR), ml	35 (0, 1,000)	10 (0, 2,000)	0.820
Ascites, median (IQR), ml	0 (0, 19)	0 (0, 50)	0.516
Duration, mean ± SD, min
Total surgery	427 ± 122	448 ± 97	0.209
Cytoreduction	292 ± 108	311 ± 84	0.206
Reconstruction	77 ± 38	78 ± 40	0.928
Splenectomy, No. (%)	7 (8.3)	5 (6.0)	0.549
Use of ureteral stent, No. (%)	52 (61.9)	55 (65.5)	0.630
Bowel resection, No. (%)	35 (41.7)	41 (48.8)	0.352
Completeness of cytoreduction, No. (%), points†			0.730
0	18 (21.4)	13 (15.5)
1	9 (10.7)	10 (11.9)
2	4 (4.8)	6 (7.1)
3	53 (63.1)	55 (65.5)
Complete cytoreduction, No. (%)‡	27 (32.1)	23 (27.4)	0.500
Use of cisplatin, No. (%)	82 (97.6)	83 (98.8)	> 0.999
Dose of cisplatin, median (IQR), mg	60 (60, 60)	60 (60, 60)	0.658
Anesthesia-related data
Duration of anesthesia, mean ± SD, min	497 ± 123	524 ± 107	0.133
Anesthetics/analgesics
Propofol, median (IQR), mg	1,240 (980, 1,438)	1,173 (1,010, 1,360)	0.469
Ciprofol, median (IQR), mg	25 (20, 28)	25 (20, 25)	0.799
cis-Atracurium, median (IQR), mg	40 (30, 49)	40 (31, 55)	0.156
Sufentanil, median (IQR), μg	30 (25, 35)	30 (25, 35)	0.995
Remifentanil, median (IQR), μg	2,480 (1,870, 2,875)	2,345 (2,021, 2,720)	0.597
Dexmedetomidine, median (IQR), μg	121 (80, 154)	132 (93, 159)	0.251
Flurbiprofen axetil, No. (%)	8 (9.5)	4 (4.8)	0.231
Volume infused/transfused
Crystalloids, median (IQR), ml	7,100 (5,750, 8,600)	6,100 (5,100, 7,525)	0.008
Artificial colloid, median (IQR), ml	500 (0, 1,000)	500 (0, 1,000)	0.566
20% albumin, median (IQR), ml	200 (200, 200)	200 (200, 200)	0.846
Blood transfusion, No. (%)	35 (41.7)	34 (40.5)	0.875
Packed RBCs, median (IQR), ml	0 (0, 400)	0 (0, 800)	0.466
Volume lost
Blood loss, median (IQR), ml	500 (150, 850)	500 (200, 800)	0.683
Urine output, median (IQR), ml	2,200 (1,800, 2,650)	1,280 (840, 1,602)	< 0.001
Urine output, median (IQR), ml ⋅ kg−1 ⋅ h−1	4.7 (4.0, 5.7)	2.5 (1.8, 3.4)	< 0.001
Fluid balance, median (IQR), ml§	5,700 (4,370, 6,920)	5,825 (4,662, 6,980)	0.598
Use of furosemide, No. (%)	77 (91.7)	43 (51.2)	< 0.001
Furosemide, median (IQR), mg	10.0 (4.5, 14.9)	2.0 (0.0, 10.0)	< 0.001
Dexamethasone, median (IQR), mg	10 (10, 10)	10 (6, 10)	0.868
Medications for hemodynamic support	79 (94.0)	80 (95.2)	0.732
Atropine, No. (%)	21 (25.0)	19 (22.6)	0.717
Atropine, median (IQR), mg	0.0 (0.0, 0.2)	0.0 (0.0, 0.0)	0.750
Norepinephrine, No. (%)	76 (90.5)	78 (92.9)	0.577
Norepinephrine, median (IQR), μg	508 (200, 944)	536 (243, 770)	0.507
Dopamine, No. (%)	21 (25.0)	24 (28.6)	0.601
Dopamine, median (IQR), mg	0.0 (0.0, 0.4)	0.0 (0.0, 4.0)	0.478
Mean MAP, median (IQR), mmHg‖	76 (74, 80) [1]	76 (69, 79) [3]	0.238
Mean heart rate, median (IQR), beats/min‖	66 (60, 70) [1]	67 (62, 72) [3]	0.222
Nasopharyngeal temperature
Highest, median (IQR), ℃	38.2 (37.8, 38.5)	38.3 (38.0, 38.8)	0.151
Mean, median (IQR), ℃	36.3 (36.0, 36.7)	36.4 (36.2, 36.9)	0.190
Lowest, median (IQR), ℃	35.6 (35.2, 36.0)	35.7 (35.4, 36.1)	0.194
Data within 7 postoperative days
Use of diuretics, No. (%)	69 (82.1)	66 (78.6)	0.560
Days using diuretics, median (IQR), days	2 (1, 3)	2 (1, 3)	> 0.999
Conduct of intraperitoneal chemotherapy, No. (%)	68 (81.0)	71 (84.5)	0.540
Cisplatin-containing chemotherapy, No. (%)	53 (63.1)	45 (53.6)	0.211

Numbers in square brackets indicate patients with missing data. P values in bold indicate values less than 0.05.

^*Used to quantify peritoneal tumor burden before surgery. The abdominopelvic cavity was divided into nine regions, and the small bowel was divided into four parts; each region or part was score from 0 to 3 according to tumor volume, resulting a total score from 0 to 39, with higher score indicating higher tumor burden.²⁶

^†Evaluated with a 4-point scale, where 0 indicates no macroscopic residual nodule, 1 indicates residual nodule less than or equal to 2.5 mm, 2 indicates residual nodule greater than 2.5 mm but less than or equal to 2.5 cm, and 3 indicates residual nodule greater than 2.5 cm.²⁶

^‡Including completeness of cytoreduction scores 0 and 1.²⁶

^§Calculated as volume infused/transfused minus volume lost.

^‖Average value from end of anesthesia induction to the end of surgery.

IQR, interquartile range; MAP, mean arterial pressure; RBC, red blood cell.

Efficacy Outcomes

The incidence of AKI within 7 days was lower in the urine-guided group than in the control group (21.4% [18 of 84] vs. 39.3% [33 of 84]; RR, 0.55; 95% CI, 0.33 to 0.89; P = 0.012). This was mainly because that the proportion of AKI diagnosed according to urine criteria (urine output less than 0.5 ml ⋅ kg⁻¹ ⋅ h⁻¹ for 6 h or longer) was less in the urine-guided group than in the control group (21.4% [18 of 84] vs. 35.7% [30 of 84]; RR, 0.60; 95% CI, 0.36 to 0.99; P = 0.040). Per-protocol analysis gave similar results (22.2% [18 of 81] vs. 39.8% [33 of 83]; RR, 0.56; 95% CI, 0.34 to 0.91; P = 0.015; table 3; supplemental table S1, https://links.lww.com/ALN/E177).

Table 3.

Postoperative Outcomes

Outcome	Urine-guided (n = 84)	Control (n = 84)	Estimated Effect (95% CI)*	P Value
Primary endpoint
Acute kidney injury (intention-to-treat), No. (%)†	18 (21.4)	33 (39.3)	RD = −17.9; RR = 0.55 (0.33, 0.89)	0.012
Diagnostic bases for acute kidney injury, No. (%)†
Urine output of <0.5 ml ⋅ kg−1 ⋅ h−1 for ≥6 h	18 (21.4)	30 (35.7)	RD = −14.3; RR = 0.60 (0.36, 0.99)	0.040
Creatinine increased ≥26.5 µmol ⋅ l−1 within 48 h	1 (1.2)	1 (1.2)	RD = 0; RR = 1.00 (0.06, 15.7)	> 0.999
Creatinine increased ≥1.5 times baseline within 7 days	1 (1.2)	3 (3.6)	RD = −2.4; RR = 0.33 (0.04, 3.14)	0.620
Acute kidney injury (per-protocol), No. (%)†	18 (22.2; n = 81)	33 (39.8; n = 83)	RD = −17.6; RR = 0.56 (0.34, 0.91)	0.015
Secondary endpoints
Stage of acute kidney injury, No. (%)†				0.006
Stage 1	17 (20.2)	23 (27.4)
Stage 2	1 (1.2)	10 (11.9)
ICU admission, No. (%)	43 (51.2)	51 (60.7)	RD = −9.5; RR = 0.84 (0.64, 1.11)	0.214
With intubation, No. (%)	42 (98.0; n = 43)	50 (97.7; n = 51)	RD = 0.3; RR = 1.00 (0.94, 1.06)	> 0.999
Duration of mechanical ventilation, median (IQR), min	243 (208, 367)	278 (234, 507)	HR = 1.41 (0.92, 2.15)	0.110
Length of ICU stay, median (IQR), min	903 (811, 981)	906 (864, 956)	HR = 1.00 (0.66, 1.51)	0.999
Other organ injuries within 7 days, No. (%)	17 (20.2)	26 (31.0)	RD = −10.8; RR = 0.65 (0.38, 1.11)	0.112
Delirium‡	1 (1.2)	8 (9.5)	RD = −8.3; RR = 0.13 (0.02, 0.98)	0.040
Myocardial injury§	7 (8.3)	10 (11.9)	RD = −3.6; RR = 0.70 (0.28, 1.75)	0.443
Acute hepatic injuries‖	10 (11.9)	11 (13.1)	RD = −1.2; RR = 0.91 (0.41, 2.03)	0.816
In-hospital stay after surgery, median (IQR), days	13 (12, 15)	14 (13, 15)	HR = 1.26 (0.93, 1.71)	0.081
Major complications within 30 days, No. (%)#	31 (36.9)	47 (56.0)	RD = −19.1; RR = 0.66 (0.47, 0.92)	0.013
All-cause 30-day mortality, No. (%)	0 (0)	0 (0)	N/A	N/A
Exploratory analyses
Time to gastric tube removal, median (IQR), days	5 (4, 7)	6 (4, 7)	HR = 1.19 (0.87, 1.61)	0.194
Time to defecation, median (IQR), days	7 (5, 9)	8 (7, 10)	HR = 1.28 (0.94, 1.73)	0.062
Length of complete intravenous nutrition, median (IQR), days	10.5 (8, 12)	11 (9, 13)	HR = 1.20 (0.88, 1.62)	0.190
Comprehensive complication index, median (IQR), points**	12.2 (0.0, 22.6)	20.9 (8.7, 29.6)	Median D = −5.4 (−8.7, 0.0)	0.009

The P values in bold indicate values less than 0.05.

^*Calculated as the urine-guided group versus or minus the control group.

^†According to Kidney Disease Improving Global Outcome creatinine or urine criteria.³²

^‡Diagnosis of incident delirium was based on established criteria according to the Confusion Assessment Method for nonintubated patients²⁷ or Confusion Assessment Method for the Intensive Care Unit for intubated patients²⁸; assessed twice daily (8 to 10 am and 6 to 8 pm) for the first 4 days and daily for the 5 to 7 days after surgery.

^§Defined as a peak high-sensitivity troponin I level exceeding the 99th percentile of upper reference limit.²⁹

^‖New-onset rise in serum total bilirubin ≥33 µmol ⋅ l⁻¹.³⁰

^#Defined as new-onset conditions that were deemed harmful and required therapeutic intervention, i.e., grade 2 or higher on Clavien–Dindo classification.³¹ Included acute kidney injury of grade 2 or higher, myocardial infarction, agitated delirium requiring restriction and/or medications, and hepatic injury requiring medications (also see supplemental table S2).

^**Calculated for each patient using the freely available online tool www.assessurgery.com according to Clavien–Dindo classifications; scores range from 0 (no postoperative complication) to 100 (death of a patient).³³

HR, hazard ratio; ICU, intensive care unit; IQR, interquartile range; N/A, not applicable; RD, rate difference; RR, relative risk.

Among secondary endpoints, patients in the urine-guided group developed less stage 1 and stage 2 AKI (stage 1, 20.2% [17 of 84]; stage 2, 1.2% [1 of 84]) than those in the control group (stage 1, 27.4% [23 of 84]; stage 2, 11.9% [10 of 84]; P = 0.006). None of our patients developed stage 3 AKI. The incidence of delirium within 7 days (1.2% [1 of 84] vs. 9.5% [8 of 84]; RR, 0.13; 95% CI, 0.02 to 0.98; P = 0.040) and the incidence of major complications within 30 days (36.9% [31 of 84] vs. 56.0% [47 of 84]; RR, 0.66; 95% CI, 0.47 to 0.92; P = 0.013) were all lower in the urine-guided group than in the control group. In exploratory analyses, the comprehensive complication index was also lower in the urine-guided group than in the control group (median 12.2 points [IQR, 0.0 to 22.6] vs. median 20.9 points [IQR, 8.7 to 29.6]; median difference, −5.4 points; 95% CI, −8.7 to 0.0; P = 0.009; table 3; supplemental table S2, https://links.lww.com/ALN/E177).

Safety Outcomes

Adverse events were comparable between the two groups. The most frequent adverse events were bradycardia (75.6%), hypothermia (69.6%), hypotension (68.5%), and tachycardia (35.7%), respectively (table 4).

Table 4.

Adverse Events

Adverse Event	Urine-guided (n = 84)	Control (n = 84)	P Value
Tachycardia, No. (%)*	30 (35.7)	30 (35.7)	> 0.999
Bradycardia, No. (%)†	67 (79.8)	60 (71.4)	0.209
Hypotension, No. (%)‡	56 (66.7)	59 (70.2)	0.618
Hypertension, No. (%)§	3 (3.6)	1 (1.2)	0.620
Hypothermia, No. (%)‖	60 (71.4)	57 (67.9)	0.615
Hyperthermia, No. (%)#	4 (4.8)	4 (4.8)	> 0.999
New-onset arrhythmia, No. (%)	0 (0.0)	0 (0.0)	> 0.999
Massive hemorrhage, No. (%)**	3 (3.6)	6 (7.1)	0.496
Desaturation, No. (%)††	0 (0.0)	1 (1.2)	> 0.999
Nausea and vomiting, No. (%)‡‡	0 (0.0)	1 (1.2)	> 0.999

Patients were monitored from the start of anesthesia induction until they left the postanesthesia care unit or 2 h after admission to the intensive care unit.

^*Heart rate of >100 beats/min or an increase of >30% from baseline.

^†Heart rate of <40 beats/min or a decrease of >30% from baseline.

^‡Systolic blood pressure of <80 mmHg or a decrease of >30% from baseline.

^§Systolic blood pressure of >180 mmHg or an increase of >30% from baseline.

^‖Nasopharyngeal temperature of <36°C.

^#Nasopharyngeal temperature of >38°C during cytoreductive surgery or >39°C during hyperthermic intraperitoneal chemotherapy phrase.

^**Blood loss of >1,500 ml or 30% of total blood volume.

^††Pulse oxygen saturation of <90% in room air.

^‡‡Any retching, vomiting, or requirement for antiemetics.

Discussion

The results of our trial showed that intraoperative urine-guided hydration targeting urine output greater than or equal to 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ or greater than or equal to 200 ml ⋅ h⁻¹ reduced postoperative AKI by more than 40% in patients who underwent CRS-HIPEC for pseudomyxoma peritonei. Urine-guided hydration was also associated with fewer major complications after surgery. The intervention strategy could be safely conducted.

AKI remains common after CRS-HIPEC, with an overall incidence of about 23.4%.¹⁰ Use of nephrotoxic chemotherapeutic agents during HIPEC is an important cause. Indeed, a recent systematic review reported an incidence of 34.7% after cisplatin-based HIPEC.¹⁰ Extensive peritonectomy and intraoperative hyperthermia might also contribute to the AKI development.^36,37 In the current study, 39.3% of our control group patients developed AKI after surgery, well within the reported range.

In previous retrospective studies, lower urine output and less fluid infusion during CRS-HIPEC were found to be important predictors of AKI development.^8–10,36 Similar association was also reported in patients undergoing other noncardiac surgeries.³⁸ Available guidelines did recommend that intraoperative urine output should be maintained at more than 1 to 2 ml ⋅ kg⁻¹ ⋅ h⁻¹ during CRS-HIPEC.^16,17 However, the optimal urine output target in these patients remains unclear.¹⁸ Furthermore, no interventional studies investigated the effect of urine-guided fluid management on the incidence of postoperative AKI in patients undergoing CRS-HIPEC.¹⁰ In the current study, we adopted a urine output target (greater than or equal to 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ or greater than or equal to 200 ml ⋅ h⁻¹) according to results in other patient populations who were at risk of AKI, i.e., patients with rhabdomyolysis, receiving iodine contrast agent, or undergoing cardiac surgery with cardiopulmonary bypass;^12–15 intraoperative fluid infusion was adjusted according to urine output and hemodynamic monitoring.

Of our patients, intraoperative urine output in the urine-guided group was maintained at (median) 4.7 ml ⋅ kg⁻¹ ⋅ h⁻¹, slightly higher than we planned; that in the control group was 2.5 ml ⋅ kg⁻¹ ⋅ h⁻¹, in line with the existing guidelines. Patients in the urine-guided group received 1 l more fluid infusion but had a similar fluid balance when compared with those in the control group. Our results showed that intraoperative urine-guided hydration reduced AKI after CRS-HIPEC by 45%. We note that 92.2% (47 of 51) of our patients who developed postoperative AKI were diagnosed according to the urine output criteria, and the between-group difference was only found in the subgroup of patients whose AKI was diagnosed according to the urine output criteria alone. This raises concern regarding the clinical significance of our findings. However, accumulating evidence showed that urine output is an early and strong predictor of AKI and that AKI diagnosed according to urine output criteria is also associated with mortality and other worse outcomes.³⁹ Further studies are warranted to confirm our findings.

Potential mechanisms underlying our findings are not totally clear but might be attributed to fluid hydration and forced diuresis. Cisplatin is mainly excreted by the kidney, but it also accumulates in kidney tissue (especially proximal tubular cells) and produces dose-dependent nephrotoxicity. Specifically, cisplatin produces direct toxic effects on cells of renal tubules, vasculature, and glomeruli and leads to tubular dysfunction and decreases in renal blood flow and glomerular filtration rate; it also induces interstitial inflammation and causes acute and chronic renal damage.^40,41 Hydration and diuresis might have reduced AKI by shortening cisplatin half-life, lowering its concentration in the urine, accelerating its transit through the proximal tubule, and thus decreasing kidney exposure to cisplatin.^42,43 Furthermore, renal ischemia is a common cause of postoperative AKI.⁴⁴ Hyperthermia per se increased the risk of AKI after CRS-HIPEC as well, possibly by inducing dehydration-related hypoperfusion and decreasing the tolerance of kidney to ischemia.³⁷ Fluid infusion is a mainstay to prevent hypovolemia and improve renal perfusion.⁴⁵Active fluid infusion might have timely replenished water loss during hyperthermia and thus guaranteed renal perfusion and reduced kidney injury.^46,47 Targeting adenosine and hypoxia signal pathways might also be helpful for kidney protection but require confirmation.^44,48

In our results, patients given urine-guided hydration less frequently developed delirium within 7 days, had fewer major complications within 30 days, had lower comprehensive complication index score after surgery when compared with those given routine hydration. We attributed these findings in the urine-guided group patients to the intervention strategy, which might have improved other outcomes by reducing AKI. Similar findings were reported in other studies.^12–15 Furthermore, urine-guided hydration could be safely conducted in our patients. However, further confirmation is required since our sample size was not estimated to detect differences in these secondary or exploratory endpoints.

There are some limitations. First, this was a single-center trial conducted only in patients with pseudomyxoma peritonei. This limits the generalizability of our results. Second, the AKI incidence in our control group patients was slightly lower than expected. This decreased the power of our trial, but we did find differences between the groups. Third, we did not perform intervention early after surgery during which period cisplatin-containing intraperitoneal chemotherapy was conducted in most patients and might have induced AKI. Fourth, the sample size was small. Results of small trials are often fragile.^49,50 Large sample-size trials are needed to verify our results.

In summary, in patients who underwent CRS and cisplatin-HIPEC for pseudomyxoma peritonei, intraoperative urine-guided hydration targeting urine output of greater than or equal to 3 ml ⋅ kg⁻¹ ⋅ h⁻¹ or greater than or equal to 200 ml ⋅ h⁻¹ reduced postoperative AKI by more than 40% and was safe. A large-scale trial is warranted to verify the effect of urine-guided hydration strategy in this patient population.

Acknowledgments

The authors gratefully acknowledge Dr. Hong-Bin Xu and his team (Department of Myxoma, Aerospace Center Hospital, Beijing, China) for help with data collection.

Research Support

Supported by National Natural Science Foundation of China (Beijing, China) grant No. 82293644 (to Dr. Wang) and National High Level Hospital Clinical Research Funding through High Quality Clinical Research Project of Peking University First Hospital (Beijing, China) No. 2022CR78 (to Dr. Wang).

Competing Interests

The authors declare no competing interests.

Reproducible Science

Full protocol available at: dxwang65@bjmu.edu.cn or wangdongxin@hotmail.com. Raw data available at: dxwang65@bjmu.edu.cn or wangdongxin@hotmail.com.

Supplemental Digital Content

Supplemental tables, https://links.lww.com/ALN/E177

Supplement Table S1. Postoperative daily therapy, creatinine, and urine output

Supplement Table S2. Individual major complications within 30 days

Supplementary Material

aln-143-1242-s001.pdf ^{(230.6KB, pdf)}