Abstract
Background/Aim: This study compared the perioperative outcomes after the same combination of hyperthermic intraperitoneal chemotherapy (HIPEC) compounds when administered for 90 min vs. 60 min, while all other therapy variables remained constant. Patients and Methods: A total of 120 patients were included with peritoneal surface malignancy who underwent cisplatin (75 mg/m2) and doxorubicin (15 mg/m2) closed HIPEC after cytoreductive surgery. Results: Sixty-five patients (54.2%) in the 60 min and 55 patients (45.8%) in the 90 min HIPEC group were compared. Patients, tumor characteristics, and postoperative complications were comparable. The only significant difference was the rate of chest drain/pleural puncture with an incidence of 18.5% and 34.5% in the 60 min and 90 min group, respectively (p=0.045). After adjustment in a multi-variable regression analysis, the odds for patients with HIPEC 90 min of having chest drain or pleural puncture in comparison to patients with HIPEC 60 min was still higher, but not significant with an OR of 2.238 (95%CI=0.932-5.373; p=0.071). Conclusion: HIPEC administered for 90 min is safe and does not increase perioperative morbidity and mortality compared to the 60-min administration.
To improve survival of patients with peritoneal carcinomatosis originating from intra-abdominal malignancies, complex multimodal therapy has been introduced over the last decades including cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC).
Surgical removal of every macroscopic tumour manifestation is not totally sufficient on its own, since the microscopic tumour cells left behind can cause recurrence (1). Since it is difficult for systemic chemotherapy to penetrate peritoneal tissues with poor blood perfusion, current practice combines cytoreductive surgery in patients with peritoneal carcinomatosis with hyperthermic intraperitoneal chemotherapy (HIPEC) in order to sufficiently treat the intraabdominal microscopic tumour tissue left behind, while systemic toxicity is minimised. There is strong evidence that hyperthermia (39-43°C) alone is tumoricidal, increases uptake and it is synergistic with chemotherapeutic agents (e.g., cisplatin, carboplatin, oxaliplatin, doxorubicin, and mitomycin C) (2-5). There has been exponential growth in the interest and publications in the field of HIPEC in the past decades, which shows increasingly wider acceptance of this multimodal therapy (1). However, while literature data are increasing, surgical factors have been the primary focus of attention and while these have been standardised worldwide, less attention has been paid to the applied HIPEC compound, and there is no standardised, widely accepted and coherent method for its most effective administration.
There is currently no systematic study assessing the optimal duration of HIPEC and the appropriate cytotoxic agent for the treatment of peritoneal metastases of different origins in human or animal models. Furthermore, based on the currently available data, it is difficult to draw a conclusion regarding the optimal duration and the influence of the duration on possible adverse effects, since the types of drugs, drug concentrations, temperatures and carrier solutions are very heterogeneous and show large variability across publications and treatment centres (1). Our goal with this study was to contribute to the determination of the optimal length of HIPEC treatment with cisplatin and doxorubicin and to quantify the effect of this therapy on the postoperative outcomes. For this, HIPEC performed in our single centre was retrospectively evaluated; the perioperative outcomes were examined and compared between the 60-min and 90-min HIPEC with cisplatin and doxorubicin therapy groups in almost identical number and characteristics of patients, while all other parameters remained constant according to the clinic intern standard.
Patients and Methods
Between August 2016 and December 2020, a total of 120 sequentially treated patients who received CRS and HIPEC with cisplatin and doxorubicin either for 60 min or 90 min in a tertiary referral centre in Germany were reviewed. The clinicopathological data had been prospectively entered in the national HIPEC registry administered by the German Society for General and Visceral Surgery (DGAV) and were retrospectively analysed for this study. Due to the retrospective nature of this study, no institutional review board approval was necessary. By providing written and verbal informed consent prior to surgery all the patients had agreed to the recording of their data in the registry and simultaneously to the use of their anonymous data for quality assurance and research purposes. All patients were treated according to multidisciplinary recommendations. Some patients were included in ongoing trials.
A closed HIPEC with a target temperature of 42°C with dual compound intraperitoneal chemotherapy with cisplatin 75 mg/m2 of body surface area (BSA) and doxorubicin 15 mg/m2 BSA was administered immediately after cytoreductive surgery to patients with peritoneal metastases of gastric or ovarian origin and mesothelioma for 60 min or 90 min. The HIPEC treatment duration was increased from 60 min to 90 min in 2018 due to an institutional protocol change. Cisplatin and doxorubicin were added to 3,000 ml-4,000 ml isotonic saline solution based on the BSA of the patients. The mean flow rate was 1,400-1,800 ml/min. During HIPEC treatments, temperatures were monitored in the right subphrenic and pelvic area.
Surgeries were performed by the same surgical team, complemented by the same gynaecological oncologist in ovarian cancer cases. The same surgeons took part in multidisciplinary patient selection process. Thus, there was only one variable difference between the two groups: the duration of HIPEC treatment. Clinicopathological data were obtained from prospectively collected database and electronic medical reports. The completeness of cytoreduction (CC) was scored as proposed by Sugarbaker: CC-0: no residual disease; CC-1: residual nodules measuring less than 2.5 mm; CC-2: residual nodules measuring between 2.5 mm and 2.5 cm; and CC-3: residual nodules larger than 2.5 cm (6).The extent of peritoneal disease was assessed by using the peritoneal cancer index (PCI) (7), which ranges from 1 to 39. Several variables (presurgical, surgical and postsurgical features) were analyzed and compared between the two groups to determine the difference in perioperative outcome and to define the most favourable HIPEC duration. Postoperative adverse events were categorized according to the Clavien-Dindo classification, and major complication was defined as ≥Grade III (8). Acute kidney injury (AKI) was classified according to the Kidney Disease – Improving Global Outcomes (KDIGO) Guideline (9), and was categorized to Stage 1 through Stage 3 depending on the serum creatinine and continuous renal replacement therapy (CRRT). Because of the heterogeneous patient groups regarding primary tumor, we did not examine recurrence-free survival and overall survival.
Statistical analysis. The distributions of continuous data are presented as mean, median, range and standard deviation. Categorical data are described using absolute frequencies and relative percentages. Metric variables with normal distribution were analyzed for differences in their means using Student’s t-test, otherwise Mann-Whitney U-test was applied. Independence of categorical variables was analyzed with Pearson’s Chi-square test; in case of small numbers, Fisher’s exact test was used.
The likelihood of experiencing a binary outcome (such as pulmonary embolism yes/no) in the comparison between patients with 90-min HIPEC and patients with 60-min HIPEC was estimated via univariable and multivariable binary logistic regression. Results are reported as odds ratios (OR). Mean differences in metric outcomes (such as length of hospital stay) between patients with 90-min and 60-min HIPEC were tested by multiple linear regression method, rendering coefficients (B) showing size and direction of the effect, i.e., mean difference.
In both methods, multivariable regression models were applied to account for unbalanced distribution of potential confounders. The following patient and tumour characteristics were considered in the multivariable models: age, sex, body-mass-index (BMI), American Society of Anesthesiologists (ASA) classification, previous chemotherapy, primary malignancy, peritoneal cancer index (PCI), and length of surgery (in min, without HIPEC). In case of the outcome chest drain/pleural puncture an additional adjustment was applied for previous peritonectomy in the right or left upper quadrant. All variables with a significance of less than p=0.100 in univariable regression analyses were selected as potential confounders in the final multivariable analyses.
Results of univariable and multivariable regressions are reported as odds ratios (OR) and regression coefficients (B), respectively, complemented by a 95% confidence interval (95%CI). Odds ratios and mean differences were considered significant if the corresponding 95%CI excluded 1 or 0, respectively. A p-value <0.05 was considered statistically significant in all tests.
The findings of this survey are presented in strict compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies (10). All analyses were performed using IBM SPSS Statistics software, Version 28.0 (IBM Corp., Armonk, NY, USA).
Results
In total, 120 patients were included in the present study. Sixty-five patients (54.2%) and fifty-five patients (45.8%) underwent 60 min and 90 min of double drug HIPEC with cisplatin and doxorubicin, respectively. Figure 1 shows the overview and distribution of the 60-min and 90-min HIPEC treatments annually.
Annual number of patients with HIPEC therapy by duration.
Demographic and clinical characteristics, i.e., age at diagnosis, sex, BMI, ASA, previous chemotherapy, location of the primary malignancy, and PCI were distributed approximately equally, and there were no significant differences in the groups of patients with HIPEC administered for 90 min and 60 min (Table I). Thirty-four patients in each group had received previous chemotherapy (52.3% of the 60-min group vs. 61.8% of the 90-min group, p=0.295). The median PCI was 10 and 11 in the 60-min and 90-min groups, respectively.
Distribution of patient and tumor characteristics by HIPEC duration.
The comparison of patients with HIPEC administered for 90 min and 60 min revealed no significant differences concerning completeness of cytoreduction (CCR) score, chemotherapy doses, length of surgery and the surgical procedures for peritonectomy, anastomoses and multivisceral resections, except for colon-rectum anastomosis (p=0.045) and small bowel resection (p=0.038), both of which were more prominent in the HIPEC 90-min group (Table II). Complete cytoreduction was achieved in 79.2% of the cases (81.8% in the 90-min group vs. 76.9% in the 60-min group, p=0.221).
Distribution of therapy by HIPEC duration.
A comparison of patients with HIPEC administered for 90 min and 60 min revealed no significant differences concerning the outcome variables depicted in Table III, except for the administration of fresh frozen plasma (36.4% in the 90-min group vs. 72.3% in the 60-min group, p<0.001) and chest drain or pleural puncture, which were more frequent in patients with HIPEC 90 min (p=0.038). Five cases of surgical infections occurred exclusively in the group with HIPEC 60 min. Based on the Clavien-Dindo classification for adverse events, 50 patients (41.7%) experienced no adverse events, whereas major, i.e., ≥Grade III complications occurred in 20% of patients (n=13) in the 60-min and 25.5% of patients (n=27) in the 90-min HIPEC groups, respectively. Revisional surgery was required in 10 cases (8.3%), with 4 patients (6.2%) and 6 patients (10.9%) in the 60-min and 90-min groups, respectively. Among the postoperative factors, acute kidney injury (AKI) occurred in 32.4% of the patients (n=39) in any stage (I-III), and no significant difference was observed between the two groups. The 30-day mortality rate was 2.5% (n=3), including two patients in the 60-min group and one patient in the 90-min group. One patient had to be resuscitated with aspiration and subsequent bradycardia with asystole, as a result of which he died in septic-toxic multiorgan failure. In the second patient, ventricular fibrillation occurred in the visceral surgery ward, the cause of which could not be clarified with certainty. Differential diagnoses primarily include a pulmonary embolism or a rhythmogenic cause in the context of QT prolongation. The third patient died of fulminant pulmonary embolism with subsequent cardiogenic shock.
Distribution of outcome variables by HIPEC duration.
After adjustment for patient and tumour characteristics in a multivariable logistic binary regression analysis, the only significant difference in binary outcome variables between the two groups remaining is the administration of fresh frozen plasma, showing a lower rate, i.e., an OR of 0.200 for the group with HIPEC 90 min compared to the group with HIPEC 60 minutes (95%CI=0.087-0.458; p<0.001, Table IV). The odds for patients with HIPEC 90 min of having chest drain or pleural puncture in comparison with patients with HIPEC 60 min is still higher, but no longer significant with an OR of 2.238 (95%CI=0.932-5.373; p=0.071).
Results from uni- and multi-variable logistic binary regression - odds ratios (OR) for outcome yes vs. no by HIPEC 90 min vs. HIPEC 60 min
No significant mean differences in length of hospital stay, length of Intensive Care Unit (ICU) stay, complication score according to Clavien-Dindo, and acute kidney injury (AKI) score could be observed in univariable or multiple regression analyses (Table V).
Results from univariable and multiple linear regression analyses for metric outcome variables: mean difference of outcome by HIPEC 90 min vs. HIPEC 60 min.
Discussion
To the best knowledge of the authors, this is the first comparative study to quantify the difference between the 60-min and 90-min HIPEC treatment groups in vivo by only changing the HIPEC therapy duration while all other variables kept constant. In this study we demonstrated that the perioperative outcome is almost identical if the duration of the HIPEC therapy with cisplatin (75 mg/m2) and doxorubicin (15 mg/m2) increased by 30 min, and the perioperative morbidity (p=0.53) and mortality (p=0.66) are not increased. There was a significant difference regarding binary outcome variables when using fresh frozen plasma (FFP) in favour of the 60-min group, and regarding the thorax drainage/pleural puncture in favour of the 90-min group. When further examining the reason for the former, we thoroughly analysed the records of the affected cases and consulted with colleagues working in the intensive care unit. We discovered that, due to a change in management, since 2018 patients have been routinely given human albumin as a plasma expander instead of FFP during intraoperative and postoperative care. Therefore, the use of fresh frozen plasma can be interpreted as a temporal change in perioperative care.
Furthermore, the increase in the incidence of intervention-worthy pleural effusion by chest drain or pleural puncture was significantly associated with peritonectomy of the upper right quadrant (p=0.006). About 83.9% of the patients with chest drain or pleural puncture had peritonectomy of the upper right quadrant as opposed to 56.2% of the patients without chest drain or pleural puncture. Concerning peritonectomy of the left upper quadrant the association was less prominent. However, the proportion of peritonectomy of the right upper quadrant under HIPEC for 90 min was only slightly higher (65.5%) than under HIPEC for 60 min (61.5%). Importantly, in multivariable analysis after adjustment for peritonectomy of the upper right and left quadrant, the association was no longer significant at p=0.073.
This finding about the increased rate of pleural effusions after peritonectomy of the diaphragm is in line with the observations of other investigators, especially if resection of the diaphragm is also necessary during the surgery (11, 12). Therefore, the higher incidence of chest drainage/pleural puncture can mostly be explained by the 30 min longer HIPEC treatment, since due to the longer HIPEC treatment more fluid can enter the chest cavity through small diaphragmatic lesions, which might remain unnoticed without HIPEC, thus causing drainage/puncture-worthy pleural effusion.
In our study the 30-day postoperative hospital mortality was 2.5%, two patients (3.1%) died in the 60-min HIPEC group and one patient (1.8%) in the 90-min HIPEC group (p=0.66). This result is in line with the German national registry, where the mortality rate was 2.3% among 2,149 consecutive patients treated with CRS and HIPEC (13).
Kusamura (14) highlights seven parameters that impact on the efficacy of HIPEC: the type of drug, drug concentrations, carrier solution, volume of the perfusate, temperature of the perfusate, treatment duration, and the technique of delivery. In addition, other authors complement these with an eighth factor: patient selection (1). The importance of the latter is underlined by the ratio of patients with CC-0 or at least CC-1 resection, since patients with larger residual tumours do not benefit from tumour resection surgeries in terms of survival, because the cytotoxic agents of the HIPEC can only penetrate to a depth of few millimetres, highlighting the importance of complete or near complete cytoreduction. In any case, the remaining tumour tissue should definitely be smaller than 2.5 mm (CCR1) (5, 15). In addition to the positive effect of surgical expertise, this ratio can be increased by effective preoperative patient selection with imaging studies and diagnostic laparoscopy in order to avoid useless laparotomy, which was highlighted by Carboni et al. in a recent retrospective study (16). In our case, the CC-0 ratio was 79.2% for all patients. In this study, only the duration of the HIPEC therapy was changed from 60 min to 90 min, while all the other variables, including the surgeons performing the surgeries, remained constant. Therefore, we would add a ninth factor to the variables, that is the surgical team, which ideally consists of surgeons proficient in cytoreductive surgery and have passed the learning curve (17).
While postoperative renal dysfunction after CRS and HIPEC is known to be multifactorial (17), increasing evidence shows that the applied cisplatin causes significant nephrotoxicity. In the study of Kusamura et al., the higher, 240mg cisplatin dose was not only associated with severe morbidity, but also increased creatinine levels (18). In our study, the highest cisplatin dose was 199.5 mg, so none of the patients reached the 240 mg limit. Most studies use the risk, injury, insufficiency, loss, and end-stage renal disease (RIFLE) (19) or the acute kidney injury network (AKIN) classification (20) for the classification of acute kidney injury. Our study used the KDIGO AKI classification for the classification of post HIPEC renal failure (9). A total of 39 patients were registered with Stage I-III deviation (32.4%), and no significant difference was found between the two groups (p=0.822). Only one patient in the 90-min group needed continuous renal replacement therapy (1.8%). For comparison, AKI development rates range from 1.3 to 40.4% based on literature data (21). In a recent study by Kurreck et al. the incidence of AKI after cisplatin-containing intraperitoneal chemotherapy in the group that did not receive sodium thiosulphate was 30.7%, which is in line with our observation. Patients in the current study did not receive the metal binding agent sodium thiosulphate during their HIPEC treatment. After results of several studies confirmed that its administration is effective in the prevention of renal failure and -is nephroprotective (22, 23), its routine administration was integrated into the protocol used in our clinic from 2021. However, it is important to emphasise that nephrotoxicity caused by cisplatin-containing HIPEC is more often reversible compared to that caused by cisplatin-containing systemic therapy (24). In the first prospective radomized multicenter study in the treatment of primary ovarian cancer, van Driel (25) and associates reported on a safe dosing of the cisplatin component of HIPEC therapy (with 50% of the dose perfused initially, 25% at 30 min, and 25% at 60 min). However, no significant difference in adverse events was demonstrated between the surgical group and the group with surgery plus HIPEC (p=0.76).
The pharmacokinetics of 60 versus 90 min HIPEC treatment is another consideration. Extensive peritonectomy during CRS has been shown not to affect the pharmacokinetics of intraperitoneal chemotherapy and thus drug transport from the peritoneal cavity to the blood plasma (26). Therefore a high concentration gradient, which exists across the peritoneal plasma barrier between the two compartments (27), can be further maintained. The dose-response curves and their dependency on exposure time have been mathematically modelled by Gardner (28). According to this model a plateau in tumour cell kill will be reached, after which prolonged exposure time offers no further cytotoxic advantage. The exposure time should be considered in relation to systemic exposure and bone marrow toxicity (29). In this context the study of Panteix et al. (30) found that the maximum platinum concentration in plasma was reached between 1 and 1.5 h after beginning the administration of cisplatin intraperitoneal hyperthermic chemotherapy.
Previous studies have confirmed that the 60-min cisplatin treatment can cause significantly higher rate of cell death than the 30-min treatment under the same conditions (1). Similarly, it can be assumed that the 90-min HIPEC treatment may have additional cytoreductive potential compared to the 60-min treatment. It is generally accepted that longer treatment causes increased exposure of the chemotherapeutic agent in the intraabdominal space; nevertheless, changes in the peritoneal clearance of the drug, as well as the thermal tolerance of the patient during prolonged therapy requires further consideration. To date, there are no systematic in vivo studies on the duration of the optimal HIPEC treatment. After quantifying the effect of the duration and its effect on adverse effects, we found that – assuming all other variables were unchanged – the safety of the 90-min HIPEC treatment was comparable to that of the 60-min treatment. In our opinion this is clinical evidence of the safety and advisability of the 90-min cisplatin and doxorubicin HIPEC treatment.
Due to the many possible variables showing large deviations in centres worldwide, and in agreement with other investigators, we also consider randomised controlled trials a less effective method for determining the contribution of the abovementioned parameters, and propose the use of in vivo and in vitro studies (1). Using the latter, the definition of the role of each variable would slowly, but gradually become available, which would contribute to the standardisation of HIPEC treatment worldwide. Other authors urge translational research and phase 1 and 2 studies for better preclinical and clinical evidence of the best possible HIPEC methodology, but in the meantime, they propose an expert consensus for better comparability of studies (31).
The limitations of this work should be addressed. The main limitation of our study is its retrospective approach and the resulting potentially inevitable bias. Besides, we did not detect long-term changes or impairment of renal function after CRS and HIPEC. Differences in survival regarding the primary tumour were not studied due to heterogeneous patient groups. Therefore, we are planning further survival evaluations in homogeneous patient groups evaluating the impact of HIPEC duration in the future. An outstanding strength of our study is that it is a comparative in vivo study, where out of all variables, only one, the duration, was changed. Thus, despite the small number of patients a valid picture could be obtained of the differences of actual perioperative outcomes. Therefore, despite the fact that this is a limited study, it provides sufficient basis for the future safe increase of the duration of treatment to 90 min in a wider patient population and larger patient groups using the abovementioned parameters as a benchmark study.
Conclusion
Since regional therapy with HIPEC is becoming more accepted and widely used, there is a need for developing a coherent and effective treatment protocol. We have determined that the 90-min treatment is effective, and – compared to the 60-minute treatment – it only increases the ratio of the pleural effusion requiring intervention, while there were no differences in the other adverse events. By only changing this single factor, the increase in the duration of the intraperitoneal exposure can safely be recommended. In order to find the most appropriate dosage and duration of the therapy, further similar comparative studies are necessary with higher number of cases, where only one out of the nine variables is modified, and the effects of every parameter are studied one at a time.
Footnotes
Authors’ Contributions
Conceived and designed the experiments: MA, LH, PP. Performed the study: MA, LH. Analysed the data: MA, LH, SI, SB, MG, BH. Contributed materials/analysis tools: BH, PP. Wrote the paper: MA, SB, PP. Designed the software used in the analysis: MG, LH.
Conflicts of Interest
All Authors have declared no potential conflicts of interest in relation to this study.
- Received December 13, 2021.
- Revision received January 4, 2022.
- Accepted January 5, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
