This study examined the impact of intraoperative GDFT based on the non-invasive PVI on GI surgical outcomes in elderly patients. The principal finding of our trial was that among elderly patients undergoing major GI surgery, PVI-directed intraoperative GDFT was not associated with a significant reduction in the proportion of patients who died or experienced complications within 30 postoperative days, compared with CFT. The analysis of secondary outcomes revealed that the intervention reduced cardiopulmonary complications but did not provide clinical benefits in terms of PONV, time to first flatus, and PLOS compared with usual fluid management.
One of the reported effects of PVI-directed GDFT was the decreased volume of fluid infusion required intraoperatively, with no effect on lactate levels (Demirel et al. 2018; Fischer et al. 2020). However, the difference in fluid administration between the GDFT and standard care groups was varied in numerous GDFT trials. A meta-analysis of 56 GDFT studies reported that the differences were within 500 mL in 35 (62%) trials, > 500 mL in 10 (18%) trials, and < 500 mL in 11 (20%) trials (Jessen et al. 2022). Greater volume difference was shown in earlier studies because of adoption of liberal fluid therapy in their control group (Sandham et al. 2003; Challand et al. 2012; Lopes et al. 2007; Benes et al. 2010). With enhanced recovery after surgery (ERAS) programmes and laparoscopic procedures being widely implemented, restricted fluid therapy has been mostly used for intraoperative conventional fluid administration, recently. In our study, the difference of total fluid administration between the CFT and GDFT groups was 425 mL, which was consistent with most findings in the literature. Nonetheless, the difference in fluid administration in the CFT and GDFT groups was not as significant as expected. This could partly explain why, in the primary analysis, GDFT based on the PVI was not found to reduce the occurrence of composite postoperative complications.
Previous studies have shown that GDFT based on different invasive haemodynamic parameters can reduce postoperative complication rates and shorten PLOS in high-risk surgery (Sandham et al. 2003; Challand et al. 2012; Lopes et al. 2007; Benes et al. 2010; Szturz et al. 2019). However, the overall certainty in the evidence was low because of the heterogeneity of the studies (Jessen et al. 2022). Our findings are not as conclusive as the results of these previous studies. We believe that the differences in the research findings can be attributed to the heterogeneous populations, varying surgeries, and inconsistencies in the implementation of fluid management protocols. In addition, with the improvement of surgical techniques and the popularisation of ERAS, the incidence of postoperative complications has decreased remarkably. Therefore, the potential effect of GDFT to reduce postoperative complications is inapparent under the current ERAS clinical path (Rollins and Lobo 2016).
Interestingly, in our secondary outcomes assessment, reduction was noted in the incidence of cardiopulmonary complications in the GDFT group compared with the CFT group (8.4% vs. 19.2%). The mechanism of the beneficial effect of GDFT could not be determined in our study. It is possible that elderly patients receiving GDFT benefit from fluid optimisation and avoid excessive fluid intake and, therefore, have a decreased risk of overload of the heart and tissue oedema, which, compared to CFT, could have potentially lowered the risk of postoperative complications such as pulmonary oedema (1.9% vs. 4.8%), arrhythmia (1.9 vs. 4.8%), pneumonia (1.9% vs. 5.8%), surgical site infection (2.8% vs. 5.8%), and anastomotic leakage (0.9% vs. 4.8%). A recent meta-analysis of GDFT trials concluded that GDFT during general anaesthesia might reduce pneumonia, surgical site infection, and anastomotic leakage, and this result reached moderate certainty in the evidence (Jessen et al. 2022). However, RCTs with greater sample sizes are warranted to accurately demonstrate statistical differences because of the low incidence of these postoperative complications.
Recovery of GI function is a significant determinant of in-hospital recovery after GI surgery (Augestad and Delaney 2010). Postoperative GI disturbance is a common complication manifested by delayed intestinal motility and PONV. Individualised GDFT guided by haemodynamic parameters seems to be the logical approach to avoid inappropriate intestinal perfusion, which can lead to postoperative GI dysfunction. In our trial, the occurrence rate of paralytic ileus and time to first flatus were similar between the two groups. GDFT was not found to improve postoperative GI function of elderly patients undergoing GI surgery. There is conflicting evidence on whether optimising fluids management can reduce the risk of PONV. In an earlier study, Gan et al. (2002) reported that GDFT guided by oesophageal Doppler monitoring results in an earlier return to bowel function and a lower incidence of PONV. In a meta-analysis, Jewer et al. (2019) concluded that supplemental intravenous crystalloid administration prevents PONV in patients undergoing surgical procedures under general anaesthesia. In our trial, there were more patients with PONV in the GDFT group (39.3%) than in the CFT group (33.7%), although this difference was not statistically significant. Large sample studies are needed to assess the risk-benefit profile of fluid therapy and PONV.
This study has some limitations. First, GDFT was performed only intraoperatively. Therefore, it is unclear if the intraoperative assessment of fluid administration is adequate to produce a considerable difference in the outcome compared to the perioperative assessment. The use of GDFT throughout the perioperative period may be more effective in improving patient outcomes. Second, the cut-off value for the PVI was a limitation. The threshold value of the PVI in our interventional group was set at 13% based on that reported in previous literature (Cannesson et al. 2008). A recent meta-analysis showed high variability regarding the best threshold value, ranging from 7–20% (Perel et al. 2014). One reason for the high variability may be differences in clinical settings and parameters of the study population, including age, vasoactive drug use, position during surgery, and pneumoperitoneum. Therefore, more studies are warranted to verify the optimum threshold value according to the study settings and participant population. Third, given the nature of GDFT, it is practically impossible to blind the clinical team performing the intervention. This may lead to potential research bias.