Pulmonary complications after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy
In this study, we found that pulmonary AEs are common after CRS and HIPEC; however, despite this, only a limited number of patients (16%) needed thoracocentesis or chest tubes. This may be a result of the strict patient selection (good performance status) for this treatment , as most of the study patients (91%) were ASA grade 1 or 2, and only two patients with pulmonary AEs had co-morbidities prior to surgery.
Patients after CRS and HIPEC treatment, usually required early post-operative ICU treatment due to influence on thoracic organs. However, the extent of treatment influence on thoracic organs is unknown. In the literature, pulmonary AEs after CRS and HIPEC have been reported previously in only two case reports, which both described the occurrence of ARDS after CRS and HIPEC [10, 14].
Peri-operative fluids and other clinical factors
To study the effect of peri-operative trauma and its effects on thoracic organs, we divided the abdomen into three surgical sites: upper, middle, and lower abdomen. We considered that the surgical trauma and the surgical site closest to the diaphragm (upper abdomen), the large amount of fluids administered during surgery, and the length of time that patients were kept on a respirator might have an influence on the development of pulmonary AEs [20, 21]. However, despite stripping of the diaphragm, the patients included in this study did not have any of the factors associated with pulmonary AEs. Therefore, insertion of a chest tube after surgery on the diaphragm should be based on individual patient signs and symptoms of AEs rather than the procedure.
Formation of atelectasis may be affected by several factors, including type and duration of anesthesia, patient position, inhaled fraction of oxygen , lack of positive end-expiratory pressure, and presence of paralysis caused by muscle relaxants [20, 22]. One study also showed that extubation failure may occur due to generous fluid treatment [23, 24]. Conversely, we did not find that the amount of fluid therapy received correlated with the occurrence of atelectasis, pleural effusion, or heart failure. However, the patients who had thoracocentesis or who had chest tubes implanted received larger amounts of crystalloids and the combination of crystalloids and colloids during surgery. None of the study patients had any signs of heart failure prior to surgery, and most of the study patients (70%) developed no signs of CHFpostoperatively. In this study, there was a weak correlation between oral intake, bowel movement, and ICU stay with the occurrence of heart failure grade 1 or 2. However, because of small number of patients with post-operative signs of congestion or CHF, it is not possible to draw any reliable conclusions.
Pulmonary AEs were not correlated with recovery parameters (restoring gastrointestinal functions and mobilization). Nevertheless, it seems that atelectasis might influence the length of post-operative hospitalization and ICU stay. Although all the patients (from our previous cohort ) were supposed to follow the same post-operative mobilization schedule  regardless of the grade of their pulmonary AE, it is possible that patients with larger atelectasis had more extensive physical therapy and mobilization than the other patients and thus stayed in the ICU for a shorter time period than patients with less atelectasis.
In our study, patients with segmental and larger atelectasis were extubated later than other groups, and patients with moderate pleural effusion had the longest hospital stay, a finding that has also been suggested in earlier studies . Conversely, good pain relief can result in patients being easier to mobilize and this might thereby result in smaller or less atelectasis. In this study, the number of patients with severe atelectasis may have been underestimated, and therefore having a larger number of patients with consecutive CT scans would have been preferable to allow us study the causality closely.
Patients in the intervention group, received larger amounts of crystalloids and the combination of colloids and crystalloids during surgery than the non-intervention group. A weight increase was seen in the entire patient population on post-operative day 1. Nevertheless, our centre’s policy is to give restricted fluid therapy during HIPEC, which is in line with other studies [6, 8, 11]. Despite good performance status, patients with higher PCI were more likely to develop pleural effusion, and this might reflect the extent of surgery. However, in this study there were no significant differences in PCI between the intervention and the non-intervention group. The intervention group also did not differ from the non-intervention group with respect to the restoration of gastrointestinal functions and mobilization.
Issues concerning radiological imaging
Radiological imaging was not routinely carried out in this study, but was performed whenever the patients status required it, for example in cases of pulmonary AE. Images were taken of only 82% of the patients, and therefore some information about the pulmonary AEs may be lacking. However, this study reflects the real-life situation, because it describes the daily care of patients with PC at our hospital. Most of the images were bedside chest radiographs, and only two CT scans were performed. It is difficult to compare the two types of image because CT scans may show more detailed findings than chest radiographs [20, 25, 26], and this could influence the findings in this study. In the future, it would be preferable to perform radiologic examinations (such as ultrasonography, CT, and chest radiography) on predetermined dates in order to be able to draw better conclusions about pulmonary AEs after CRS and HIPEC. The benefits of these two imaging methods for patients undergoing major surgery could not be addressed in the current study.
This study, assessed the incidence of pulmonary AEs after CRS and HIPEC and their effect on early recovery. Although the gradations of atelectasis, pleural effusion and heart failure have not been established, we found similar results in the literature [20, 27]. However, the radiologists who performed the gradations were blinded to the post-operative course of the study, and Cohen’s weighted κ score indicated only a moderate level of agreement between the two radiologists, which is a weakness in the study, but still demonstrates that the grading process was rigorous.