This study demonstrates that a combination of preoperative IVC and pulmonary ultrasound is a feasible way of guiding preprocedural fluid therapy, and it is an effective way to decrease the incidence of early intraoperative hypotension. Of note, the occurrence rate of relative hypotension was relatively high, but the absolute MAP criterion was diagnosed in a rate which was comparable to the results of earlier studies in the CFT group, and we were able to decrease it to lower levels [4, 13, 16]. As MAP < 65 mmHg is a potentially better surrogate of hypoperfusion, we assume that our protocol enabled us to promote patient safety [3], but a further evaluation of hard endpoints is needed.
We provided evidence that our two-step approach is safe, since it did not lead to hazardous fluid overload – as shown in the similar results in both groups’ postoperative P/F ratios and lung ultrasound scores. The observation of bilateral anterior B-profile, a safety limit in our protocol, necessitated the interruption of fluid loading only once. The beneficial effects were achieved by infusing significantly more fluid preoperatively in the treatment group, but it basically meant an earlier administration of a higher proportion of the operative day’s whole dose, which was similar in both groups.
Although all anaesthetists are aware of the well described hazards of intraoperative hypotension, this complication is still an important issue. Despite recent advances in the implementation of enhanced recovery (ERAS) protocols [31, 32], dehydration and hypovolemia can still be common before surgery and they are difficult to diagnose [33, 34]. As the aforementioned guidelines promote restrictive intra- and postoperative fluid therapy, and in the same time alerts anaesthesiologist of the detrimental effects of the liberal use of vasopressors, it is vital to identify patients at risk before procedures. Point-of-care ultrasound was verified as a promising non-invasive method in several studies, where results offered pointing out patients with a potentially modifiable risk [13, 15, 16, 18]. Despite its obvious limitations (e.g., respiratory efforts, right ventricular dysfunction, pericardial disease, higher intraabdominal pressures) [35], the measurement of IVCCI was validated as a tool to guide fluid load before subarachnoid blockade and it effectively helped prevent hypotension [17, 18]. Similar studies for general anaesthesia have not been available yet, possibly due to issues of standardisation.
Our study has some obvious limitations. The beneficial effect of USP-guided treatment seemed time dependent, as the between-group differences were more prominent at the immediate postinduction timepoint. It could be, at least partly, explained by the lack of standardisation in the maintenance of anaesthesia and by the variability of the exact start of surgical manipulation (including patient positioning). The timeframe for registering intraoperative hypotension was 12 minutes at most. This allowed us not to interfere with surgical manipulations or positioning and we did not have to postpone the start of the surgical intervention. A similar sized study of from Myrberg et al., where preoperative fluid bolus was not linked to ultrasound findings, seemed similarly effective [36], but the option of individualising the fluid loading in our protocol is an additive potential. A second limitation was accepted in order not to create a cumbersome protocol, and we limited the occasions of preoperative ultrasounds to two. This allowed for an easy-to-intervene evaluation, but some of the patients with subclinical hypovolemia probably remained untreated.
Our secondary outcome measures were appropriate to verify the safety of our USP, but we did not detect better early postoperative organ functions. These differences are also dependent on the treatment standards of the ‘control’ group. The potentially detrimental outcomes, like postoperative oliguria or lactic acidosis were rare in both study groups – possibly because we used fluid therapy of similar overall amounts on the operative day. A well performed interventional study aiming to decrease the risk of renal failure by preoperative hydration used a much higher amount of infusion than we did, and also failed to verify the effectiveness of their protocol [37]. The daily intravenous fluid regimens in our study can be classified ‘liberal’ as defined by a meta-analysis from Varadhan et al. (> 2.75 litres per day) [38], but absolute limits of fluid therapy are less useful to distinguish restrictive and liberal protocols, especially when less attention was paid to the estimation of preoperative dehydration in some previous studies. A similar ‘liberal’ therapy was not associated with a lower rate of disability-free survival and even decreased the risk of renal failure compared to a ‘restrictive’ protocol [34].