Effect of goal-directed fluid therapy based on both stroke volume variation and delta stroke volume on the incidence of composite postoperative complications among individuals undergoing meningioma resection

To the Editor: In neurosurgery, intraoperative fluid administration is essential in maintaining appropriate systemic and cerebral perfusion pressure, oxygenation, and metabolism. Meningioma resection can involve massive intraoperative blood loss due to an abundant blood supply in structures. Moreover, the routine use of mannitol during meningioma resection may affect hemodynamic stability. Fluid overload is correlated to neurological, pulmonary, or gastrointestinal (GI) edema, which has unfavorable outcomes.^[1] Thus, neurosurgeries, particularly meningioma resection, require individualized fluid therapy to achieve a balance between inadequate fluid and fluid overload.

In recent years, the concept of perioperative fluid administration has been changed from a liberal or restrictive strategy to a goal-directed fluid therapy (GDFT). Increasing evidence has shown that perioperative GDFT is correlated to accelerated recovery.^[2]

Therefore, we conducted a single-center, single-blinded study to investigate the effects of GDFT based on stroke volume variation (SVV) and delta stroke volume (ΔSV) using the LiDCO_rapid system (LiDCO Ltd., London, UK) on postoperative outcomes in individuals undergoing meningioma resection with the intraoperative use of mannitol.

The patients were randomly allocated to either a goal-directed fluid therapy group (GDFT group) or a control group at a ratio of 1:1. The study was approved by the research Ethics Committee of Xuanwu Hospital, Capital Medical University (No. LYS-2016[38]) and registered prior to patient enrollment at ClinicalTrials.gov (ChiCTR-IOR-16009007; Principle Investigator: T.L.W; date of registration: Aug 10, 2016). Written informed consent was obtained from all participants. Patients aged >18 years who had undergone elective meningioma resection with the intraoperative use of mannitol from November 2016 to July 2018 were included [Supplementary Figure 1, http://links.lww.com/CM9/B496].

Before the induction of anesthesia, a 20-gauge radial arterial line was inserted and connected to the LiDCO_rapid system (LiDCO Ltd.) to obtain the SVV and cardiac index. Patients in the control group was maintained with fluid infused at 2 mL·kg^-1·h^-1. Fluid was given according to the anesthesiologists' experience to maintain a central venous pressure (CVP) ≥6 mmHg and a mean arterial pressure (MAP) ≥80% of the baseline value. In the GDFT group, maintenance fluid was infused at a same rate as in the control group. When the measured SVV was >13% and lasted for at least 1 min, a fluid challenge was performed with 3 mL/kg of Lactated Ringer's (LR) solution to observe whether the ΔSV was >10%, thereby indicating a positive response. If a positive response was obtained after the first fluid challenge, then an additional 3 mL/kg of LR can be used until the ΔSV was <10% [Supplementary Figure 2, http://links.lww.com/CM9/B496]. I.v. ephedrine and phenylephrine were administered to maintain the targeted blood pressure according to patients' heart rate. Finally, the same team of neurosurgeons treated all patients intraoperatively and postoperatively.

The primary outcome included composite complication rate. We defined the composite complication as more than one complication occurring simultaneously in a patient. Complications were pre-defined based on the criteria of the International Statistical Classification of Diseases and Related Health Problems. The diagnosis of severe encephaledema is based on the symptoms and imaging examination. The researcher who was in charge of follow up was blinded to the study design and grouping.

Data were analyzed using the Statistical Package for the Social Sciences software version 18.0 (SPSS InC., Chicago, IL, USA). Before conducting data analysis, a normality test was conducted. A normal distribution was presented with mean ± standard deviation. The Fisher's exact test or χ² test was used to analyze dichotomous data, and the student's t-test was used for normally distributed continuous data. Moreover, the Mann–Whitney U test was used for non-parametric ordinal data. A P value <0.05 was considered statistically significant.The sample size was calculated according to the results of a pilot study about postoperative composite complication rates. According to a two-tailed power analysis with an α of 5% and β of 10%, at least 39 patients are required per group. To allow a 10% drop-out rate during the follow-up period, we planned to include 44 patients in each group.

In this study, 105 patients who underwent meningioma resection were initially included. A total of 88 patients were finally enrolled in this study, and 84 patients completed the study protocol (n = 42, GDFT group and n = 42, control group; Supplementary Figure 1, http://links.lww.com/CM9/B496). The patients in the two groups had comparable baseline characteristics and past medical history [Supplementary Table 1, http://links.lww.com/CM9/B496]. The baseline characteristics of the patients, previous medical history, and surgical information were similar between the two groups. Additionally, no significant differences were observed in terms of hemoglobin level after surgery in two groups (P >0.05).

Supplementary Table 2, http://links.lww.com/CM9/B496 shows the intraoperative profiles. The use of a higher amount of crystalloid was observed in patients treated with GDFT than the control group (2435 ± 534 mL vs. 2150 ± 592 mL; P = 0.023. As for the intraoperative hemodynamic profiles, the intraoperative averages of cardiac output (CO) and stroke volume (SV) were similar between the two groups (4.9 [range: 4.4–5.3] L/min vs. 4.6 [4.3–4.9] L/min, P = 0.153; 75 ± 14 mL vs. 71 ± 9 mL, P = 0.176). No significant differences were observed in terms of weighted area under the curve (AUC) and AUC (among CO, SV, and MAP).

The composite postoperative complication rate between the two groups is shown in Table 1. The proportion of patients who developed more than one postoperative complication were significantly smaller in the GDFT group than in the control group (69%[29/42] vs. 88%[37/42], P = 0.033). Proportion of patients who developed ≥2 and ≥3 postoperative complication were also smaller in GDFT group than in the control group (38%[16/42] vs. 62%[26/42], P = 0.029; 12%[5/42] vs. 31% [13/42], P = 0.033). The postoperative complications classified based on the Clavien-Dindo classification system (CDCS) are presented in Supplementary Table 3, http://links.lww.com/CM9/B496. The number of patients with ≥ grade II classification was significantly higher in the control group than in the GDFT group (31%(13/42) vs. 55%(23/42), P = 0.029).

Table 1.

Hospital course and postoperative complication of patients undergoing meningioma resection.

Variables	GDFT (n = 42)	Control (n = 42)	Statistics values	P values
Postoperative complications
Composite complications
≥1	29 (69)	37 (88)	4.525*	0.033
≥2	16 (38)	26 (62)	4.762*	0.029
≥3	5 (12)	13 (31)	4.525*	0.033
Overall complications
Hypertension,	7 (17)	5 (12)	0.389*	0.533
Myocardial ischemia or infarction	0	2 (5)	‡	0.494
Intubation >24 h	0	2 (5)	‡	0.494
ALI or ARDS	0	0		NA
Respiratory infection	4 (10)	7 (17)	0.418*	0.518
Stroke	0	0		NA
Intracranial hemorrhage	1 (2)	0	‡	1.000
Intracranial infection	2 (5)	1 (2)	‡	1.000
Hemiplegia	2 (5)	3 (7)	‡	1.000
Aphasia	1 (2)	0	‡	1.000
Severe encephaledema	2 (5)	11 (26)	5.824*	0.016
Constipation	19 (45)	22 (52)	0.429*	0.513
PONV	16 (38)	19 (45)	0.441*	0.507
Gastrointestinal hemorrhage	0	0		NA
Ileus	0	1 (2)	‡	1.000
Urinary infection	0	0		NA
Creatinine elevation	0	1 (2)	‡	1.000
Thrombocytopenia (Platelet <100,000)	0	3 (7)	‡	0.241
Coagulopathy (INR >1.5)	0	1 (2)	‡	1.000
Wound infection	0	0		NA
Hospital course
Length of hospital stay (days)	14 (13–15)	15 (12–17)	–0.966†	0.334
Number of ICU admission	8 (19)	7 (17)	0.081*	0.776
Get out of bed to walk (days)	2 (2–3)	3 (2–4)	–1.481†	0.139
Postoperative GI function
Postoperative exhaust time (h)	14 (5–24)	20 (11–29)	–2.070†	0.038
Take solid food (days)	2 (1–3)	3 (2–5)	–3.209†	0.001

Data are presented as median (interquartile range) for continuous variables. Data are presented as n (%) for binary variables. ^*χ² value; ^†Z value; ^‡Fisher's exact test. ALI: Acute lung injury; ARDS: Adult respiratory distress syndrome; GI: Gastrointestinal; GDFT: Goal-directed fluid therapy; ICU: Intensive care unit; INR: International normalized ratio; NA: Not available; PONV: Postoperative nausea and vomiting.

The overall postoperative complication rate is also presented in Table 1. The incidence rate of severe encephaledema after surgery was lower in the GDFT group than that in the control group (2 [5%] vs. 11 [26%], P = 0.016). Table 1 also depicts other postoperative outcomes. There was no significant reduction in terms of length of stay (LOS), number of intensive care unit (ICU) admission, and time of ambulation after surgery. As for GI function recovery, the time from surgery to the first flatus (14 [5–24] h vs. 20 [11–29] h, P = 0.038) and tolerance to solid food (2 [1–3] days vs. 3 [2–5] days, P = 0.001) were shorter in the GDFT group than in the control group.

In neurosurgical patients, fluid imbalance may contribute to the development of postoperative complications.^[3] As an optimized fluid therapy based on hemodynamic parameters, GDFT decreases the mortality, hospital length of stay, and several postoperative complications. We developed a GDFT protocol for neurosurgical patients targeted at individualized fluid administration and adequate organ perfusion, and this treatment eventually reduced the total incidence rate of postoperative composite complications.

After surgery, in patients with intracranial tumors, including meningioma, neurosurgeons are primarily concerned about peritumoral edema. In patients undergoing a major surgery, less amount of fluid was infused in the GDFT group than in the control group.^[4] This finding was different from our observation. Strict restrictive fluid therapy strategies were employed in the control group according to the anesthesiologists' experience to reduce postoperative brain edema. Similar as Wu et al's^[5] study, our data also suggested that fluid therapy targeting a lower SVV were more beneficial than a restrictive protocol.

As a significant part of enhanced recovery after surgery, accelerating the recovery of GI function might prevent intestinal flora imbalance, reduce the rate of postoperative infection, and even shorten LOS. Recently, GDFT was confirmed to facilitate GI function recovery,^[6] and these results were consistent with those of our study.

For treatment innovations, we designed a comprehensive fluid therapy protocol based on SVV and ΔSV, which is different from that used in the protocol of other previous GDFT trials. This study highlighted a GDFT method that can be used in patients undergoing meningioma resection using the LiDCO_rapid system. This approach is minimally invasive and easy to use. Moreover, it can facilitate continuous real-time monitoring.

The current study had some limitations. First, this was a single-center and single-blinded study. Patients with meningioma were included. However, the inclusion was not restricted based on the size or location of the tumor. This study was underpowered for the detection of long-term mortality and morbidity. Second, we excluded patients with body mass index (BMI) with extreme ranges (according to Asian standards) to minimize bias on SVV validity due to changes in thoracic compliance. Nevertheless, the value of the GDFT protocol may be helpful for patients with a larger weight with a different cut-off value for functional hemodynamic parameters. Third, the postoperative hemodynamic parameters were only monitored but not recorded for comparison.

In conclusion, the GDFT protocol based on SVV and ΔSV may be more beneficial in patients undergoing meningioma resection with the intraoperative use of mannitol than in those who received the conventional fluid therapy.

Conflicts of interest

None.

Supplementary Material

cm9-136-1990-s001.pdf ^{(455.2KB, pdf)}