Cytoreductive surgery with hyperthermic intraperitoneal chemotherapy – perioperative management at Waikato Hospital

Dear Editor,

We read with interest the article by Shiralkar et al. with regard to the anaesthetic management of patients undergoing cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) at the Princess Alexandra Hospital in Brisbane, Queensland. ¹

Waikato Hospital is the only public hospital in New Zealand providing cytoreductive surgery and HIPEC. New Zealand’s population is only slightly less than that of Queensland, so we would expect to see relatively similar volumes of surgery. Surgical technique is similar to that described by Shiralkar et al., with the notable exception that post-operative chemotherapy is not undertaken. ² However, being a relatively resource-limited health care system compared with Australia, our anaesthetic technique has been adapted to limit the resource burden of this complex operation.

Our experience over the past decade from 2008 to 2017 includes 76 surgeries in 72 patients where HIPEC took place, with four having repeat procedures for tumour recurrence. Patients who came forward for surgery during this time period but had irresectable disease were excluded from this analysis. The median age was 55 years with a range of 25 to 74 years. We have not recorded the indications for surgery, but from previously published work the majority of cases were pseudomyxoma peritonei. ² All patients received mitomycin C as the chemotherapeutic agent during HIPEC, and four patients additionally received cisplatin for intra-abdominal mesothelioma. There were no intraoperative deaths, but one patient (1.3%) died during their hospital admission.

Our anaesthetic technique differs significantly from that described by Shiralkar et al. in three main ways: the near routine use of epidural analgesia for these patients (96% of patients had a pre-operative epidural); the conscious effort to limit fluids; and the use of vasopressin to reduce splanchnic blood flow. While 21 different anaesthetists anaesthetised these patients during the time period, two anaesthetists were responsible for anaesthetising half of the cases.

When comparing fluid administration to the data provided by Shiralkar et al., our median crystalloid administration was 5000 ml (range 2000–20,000 ml) and our median total fluid administration excluding blood products was 5300 ml (range 2000–22,000 ml), indicating very little colloid use. Only three patients (4%) received more than 15,000 ml of fluid intra-operatively, compared with 17 patients (24%) in the report by Shiralkar et al. This combined with near routine use of epidural analgesia allowed us to extubate 73 of 76 patients at the end of the procedure, compared with three patients extubated at the end of the procedure by Shiralkar et al. Of these 76 patients, 70 went to a high dependency unit, five went to the Intensive Care Unit (ICU) (including the three who were not extubated), and one patient went directly to a ward.

The use of blood products in our cohort was also significantly less than that reported by Shiralkar et al.; 34% of our patients received packed red blood cells (PRBC), 12% received fresh frozen plasma (FFP), 6.5% received platelets and 4% received cryoprecipitate. A simple chi-squared analysis reveals that our patients received statistically significantly less PRBC, FFP and cryoprecipitate than the Brisbane cohort. The difference in platelet transfusion was not statistically significant.

Vasopressors were used at the discretion of the treating anaesthetist. In every operation, a peripheral vasopressor, either metaraminol or phenylephrine, was used. Vasopressin was also used in 58% of cases and Noradrenaline was used in 14 cases (18%). Eighteen patients (24%) received tranexamic acid intra-operatively. We do not have enough data to investigate whether deep vein thrombosis was more common in this group than that seen by Shiralkar et al.

Possibly as a result of the relatively restrictive fluid regimen, our rates of acute kidney injury were higher than that observed by Shiralkar et al., with eight of 76 (10.5%) having a rise in creatinine >1.5 times baseline within the first five days of surgery. However, no patient required dialysis.

Using the presence of a platelet count <100 × 10⁹/L, international normalized ratio (INR) ≥1.5, activated partial thromboplastin time (APTT) ≥45 s to define coagulopathy and a platelet count of <50 × 10⁹/L, INR >2.0, APTT >60 s to define severe coagulopathy, we found rates of 26% and 4%, respectively. Coagulopathy has been investigated in other centres offering CRS where epidurals are commonly used; identified risk factors include intraoperative transfusion and increased peritoneal carcinomatosis index (an assessment of tumour burden). ³ In previous studies, the coagulopathy frequently resolves by day four post-operative and a transfusion to correct this abnormality prior to planned epidural removal is uncommon.^3,4

Overall, we found significantly smaller volumes of intra-operative fluid administration and a higher rate of post-operative acute kidney injury compared with Shiralkar et al. These findings may be linked given the results of the recently published restrictive versus liberal fluid therapy in major abdominal surgery trial. ⁵ Peritonectomy surgery has been associated with significant post-operative pain, chronic pain and poor quality of life. ⁴ In an international survey, 72% of centres practicing CRS with HIPEC routinely used thoracic epidurals. ⁶ They have been shown to be effective in managing pain after peritonectomy surgery and associated with higher rates of successful extubation in theatre or shorter duration of ventilation in ICU. ⁷ Transient post-operative coagulopathy is not uncommon but has not been shown to complicate epidural use, when coagulation studies are measured prior to removal. ⁴ While post-operative ventilation in ICU is commonly practised after CRS with HIPEC, we have found that almost routine extubation is possible, reducing the burden on critical care.