Comparison of early recovery and cognitive function after desflurane and sevoflurane anaesthesia in elderly patients: A meta-analysis of randomized controlled trials

Abstract

Objectives

A meta-analysis to compare postoperative cognitive function and the time to specific recovery events in elderly patients (aged >65 years) anaesthetized with sevoflurane or desflurane.

Methods

A systematic search of the PubMed®, Embase®, Cochrane Library and Chinese Biomedical databases was performed using the keywords ‘sevoflurane’ and ‘desflurane’. Data and characteristics of appropriate randomized controlled trials (RCTs) were extracted.

Results

The meta-analysis included five trials (n = 300). The time taken to follow commands (mean difference [MD] −3.27; 95% confidence intervals [CI] −4.95, −1.59), extubation (MD −1.59; 95%CI −2.62, −0.55), orientation (MD −4.31; 95%CI −4.99, −3.62), and recovery room discharge (MD −9.38; 95%CI −13.43, −5.42) were significantly shorter in the desflurane group than in the sevoflurane group. There was no significant between-drug difference in the incidence of postoperative cognitive dysfunction or the time taken to open the eyes.

Conclusions

Desflurane is associated with a faster recovery from general anaesthesia than sevoflurane in elderly patients.

Introduction

Elderly individuals (those aged >65 years) account for an increasing proportion of medical provision globally, due to economic development and extended life expectancy. ¹ In 2013, 17% of the population of developed countries was aged >65 years and this is expected to rise to 18.1% by 2050. ² Because of age-related degeneration in respiratory and cardiovascular function and other associated systemic diseases, elderly patients are at risk of experiencing anaesthesia complications. Postoperative cognitive dysfunction (POCD) was found to be present in 25.8% of elderly patients at 1 week after surgery, and 9.9% of elderly patients at 3 months after surgery. ³ In addition, elderly patients who developed POCD were twice as likely to develop dementia in the following 3–7 years after anaesthesia and surgery compared with those unaffected by POCD. ⁴

The pathogenesis of POCD is not clear, but risk factors include old age, pre-existing cerebral, cardiac or vascular disease, alcohol abuse, and intra- and postoperative complications. ⁵ Anaesthetic drugs can also affect postoperative cognition, as their residual effects can alter central nervous system activity.^6,7 Rapid recovery from anaesthesia may therefore reduce the incidence of many postoperative complications, such as POCD and postoperative respiratory distress.

Desflurane and sevoflurane, which are the most widely used inhalation anaesthetics in the elderly, ⁸ are both characterized by a low blood–gas partition coefficient, and allow more rapid recovery and earlier discharge than traditional volatile anaesthetics.^9–11 Some studies have shown that patients receiving desflurane were quicker to follow commands, and to be extubated and oriented, than patients receiving sevoflurane;^12–14 however, no difference in recovery time was found in elderly patients assessed using the Digit–Symbol Substitution Test. ¹⁵ In addition, although some studies found a similar incidence of POCD following desflurane or sevoflurane anaesthesia,^12–14 others demonstrated that desflurane anaesthesia resulted in better postoperative cognitive function than sevoflurane anaesthesia in elderly patients.^16,17

The aim of the present meta-analysis was to compare postoperative cognitive function and the time to specific recovery events in elderly patients anaesthetized with sevoflurane or desflurane.

Materials and methods

Inclusion criteria

The meta-analysis included randomized controlled trails of elderly patients (aged >65 years) who received either sevoflurane or desflurane for anaesthesia maintenance. Data representing recovery profile, (including time to open eyes, follow commands, extubation, orientation and discharge from recovery room) must be presented as the mean and a measure of variance. Data representing side-effects or complications must be presented as n, to allow pooling of data.

Search strategy

PubMed®, Embase®, Cochrane Library and Chinese Biomedical databases were searched from date of inception to January 2015 using the keywords ‘sevoflurane’ and ‘desflurane’ in combination. Results were limited to randomized controlled trials in humans, and in the elderly (>65 years) patient population. Articles were identified by title and abstract. If the intervention or outcome indicator was inappropriate, the study was excluded. The full text of the remaining abstracts were sought and assessed according to predetermined criteria. Reference lists of other related systematic reviews and literature acquired from online searching were also examined for any missing articles.

Data extraction

Data were extracted independently by two investigators (G.C. and Y.Z.), with disagreements resolved by consensus. Data included patient characteristics, anaesthesia strategies, time to open eyes, follow commands, extubation, orientation and discharge from the recovery room, and the incidence of POCD. Since studies used different terminology to describe recovery indicators, consensus was reached that ‘spatial orientation’ was equivalent to ‘provide birth date clearly’, and ‘follow commands’ was equivalent to ‘squeeze fingers’.

Assessment of methodological quality

Assessment of risk of bias for included studies was made by two independent investigators (Q.S. and H.Z.) according to Cochrane collaboration guidelines ¹⁸ that comprise six items: random sequence generation, allocation concealment, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias. Each item was assessed as low risk, unclear risk or high risk. If agreement could not be reached, a third investigator (Y.Z.) made the final decision.

Statistical analyses

Extracted data were analysed using RevMan version 5.2 (Cochrane Collaboration, London, UK). Mean difference (MD) was calculated for continuous variables, and odds ratio (OR) was calculated for dichotomous variables, both with corresponding 95% confidence intervals (CI). The inverse variance method was used to pool MDs, and ORs were pooled using the Mantel–Haenszel method. Heterogeneity among studies was determined using χ²-test and I²-value. A fixed-effects model was used to combine results in the case of significant heterogeneity (P < 0.05 or I²> 50%); otherwise a random-effects model was used. Subgroup analysis was performed to combine the incidence of POCD at different time points after surgery.

In some studies, continuous variables were presented as mean with 95% CI. The following formula was used to estimate SD when the data conforms to normal distribution: N ≤ 60, SD =  $\sqrt{N}$  × (upper 95% CI − lower 95% CI)/(2 × t), t = tinv (probability, deg-freedom); N ≥ 100, SD =  $\sqrt{N}$  × (upper 95% CI − lower 95% CI)/3.29; 60 < N < 100, either formula is available.

Results

The initial literature search identified 435 articles. After exclusion of irrelevant articles, 37 studies were assessed in detail, five of which met the inclusion criteria (age range 65–88 years; total n = 300; desflurane n = 150; sevoflurane n = 150).^{12–14,19,20} Details of the study selection procedure are shown in Figure 1 and the characteristics of the included studies are described in Table 1. The methodological quality of the studies is presented in Figure 2.

Figure 1.

Flow chart of the study selection procedure in a meta-analysis to compare postoperative cognitive function and the time to specific recovery events in elderly patients (aged >65 years) anaesthetized with sevoflurane or desflurane.

Figure 2.

Summary of risk of bias in studies included in the meta-analysis.

Table 1.

Characteristics of randomized controlled studies included in a meta-analysis to compare early recovery and cognitive function after desflurane or sevoflurane anaesthesia in elderly patients.

Author, year of publication	ASA	Type of surgery	N2O used	Volatile anaesthetic, MAC/h or %	Age, y	Sex, M/F	Total n	Anaesthesia duration, min	Cognitive function scale
Chen et al, 2001 14	I–III	TKR/THR	Yes	Des 2.5 ± 0.6 Sev 2.7 ± 0.5	75 ± 8 73 ± 9	20/15 18/17	35 35	155 ± 42 161 ± 39	MMSE
Iannuzzi et al, 2005 19	II	Mixed	Yes	Des 1.92 ± 1.04 Sev 1.66 ± 1.49	71 ± 3.8 70 ± 4.2	5/9 15/7	18 18	64 ± 10.3 68 ± 8.4	NR
Rortgen et al, 2010 13	I–III	Mixed	No	Des 4.3 ± 0.26% Sev 1.2 ± 0.11%	65–75 65–75	NR	40 40	127 (105–149) 152 (133–170)	TAP, PPT
Zhang, 2010 20	II	LC	Yes	Des 1.97 ± 1.03 Sev 1.75 ± 1.32	71 ± 5.6 69 ± 5.2	16/11 18/9	27 27	69 ± 10.3 66 ± 9.6	NR
Deepak et al, 2013 12	I–III	Mixed	Yes	Des 1–1.8 Sev 1–1.8	69.1 ± 4.7 69.4 ± 4.4	12/18 19/11	30 30	145 ± 40.77 141 ± 35.26	MMSE

ASA, American Society of Anesthesiologists physical status scale; Des, desflurane; Sev, sevoflurane; NR, not reported; TAP, attentional performance test; PPT, classical paper–pencil test; TKR, total knee replacement; THR, total hip replacement; MMSE, mini-mental state examination; ²¹ LC, laparoscopic cholecystectomy.

There was no significant between-drug difference in the time taken to open eyes (five studies, n = 300^{12–14,19,20}; MD −0.82; 95% CI −2.42, 0.78; I²= 88%, P < 0.01; Figure 3A). The time taken to follow commands was significantly shorter in the desflurane group than the sevoflurane group (three studies, n = 210^12–14; MD −3.27; 95% CI −4.95, −1.59; I²= 68%, P = 0.05; Figure 3B). Time to extubation was significantly shorter in the desflurane group than the sevoflurane group (five studies, n = 300^{12–14,19,20}; MD −1.59; 95%CI −2.62, −0.55; I²= 79%, P < 0.01; Figure 3C). Patients receiving desflurane oriented significantly sooner than those receiving sevoflurane (four studies, n = 246^12–14,19; MD −4.31; 95%CI −4.99, −3.62; I²= 0%; P = 0.4; Figure 3D). Patients receiving desflurane also met recovery room discharge criteria significantly sooner than those receiving sevoflurane (four studies, n = 240^13,14,19,20; MD −9.38; 95%CI −13.43, −5.42; I²= 0%, P = 0.49; Figure 3E).

Figure 3.

Meta-analysis of the time to (A) open eyes, (B) follow commands, (C) extubation, (D) orientation, and (E) recovery room discharge after desflurane or sevoflurane anaesthesia in elderly patients (>65 years of age).

Postoperative cognitive dysfunction was assessed in four studies.^12–14,20 A variety of clinical tools were used and the timing of measurements differed widely between these studies. The Mini-Mental State Examination (MMSE) ²¹ was used in two studies (n = 150),^12,14 and meta-analysis found no significant between-group difference at 1 h, 3 h and 6 h postoperatively (Figure 4). Subgroup analysis of all POCD data found no significant between-group difference in cognitive function (n = 620; OR 0.96; 95%CI 0.64, 1.44; P = 0.83; I²= 0%; Figure 4).

Figure 4.

Meta-analysis of the incidence of postoperative cognitive decline after desflurane or sevoflurane anaesthesia in elderly patients (>65 years of age).

Discussion

The present meta-analysis found that the time to several recovery events was significantly shorter following desflurane than sevoflurane anaesthesia in elderly patients, including following commands, extubation, orientation and discharge from the recovery room. There were no significant between-group differences in postoperative cognitive function or the time taken to open the eyes following anaesthesia, however.

The quicker recovery characteristics of desflurane versus sevoflurane identified by our meta-analysis are likely due to the very low lipid solubility of desflurane. ²² The speed of tissue wash-in and wash-out is determined by tissue/blood partition coefficient, and the blood/brain partition coefficients of desflurane and sevoflurane are 1.29 ± 0.05 and 1.7 ± 0.09, respectively. ²² The time taken to open eyes is a widely used indicator for early recovery from anaesthesia, but there was no significant between-group difference in this parameter in the present study. This may be explained by the concurrent use of different drugs in the included studies, since the residual effects of sedatives, analgesics and muscle relaxants may affect the onset of early recovery. In addition, the included studies did not use standardized methods to measure the depth of anaesthesia, and patients can be ‘fast tracked’ through recovery when bispectral index (BIS) is used to monitor anaesthetic depth. ²³ The depth of anaesthesia was assessed based on haemodynamic changes in two of the studies included in our meta-analysis,^12,13 which would have complicated attempts to maintain similar anaesthetic depth in the two study groups.

Several factors have been shown to contribute to POCD, including increasing age, alcohol abuse, hypoxia, hypotension and type of surgery. ²⁴ The pooled estimate of the OR of POCD from our meta-analysis found no difference between sevoflurane and desflurane anaesthesia in elderly patients, but this conclusion should be treated prudently because of the small numbers of studies included in this part of the analysis. Because the incidence of POCD is low, a large patient cohort is required in order to achieve statistically significant findings. The variety of clinical tools used to measure cognitive function and the lack of a clear definition for POCD makes it difficult to compare data.

The present study has several limitations. The analysis included only five studies, while three additional related articles were excluded due to inappropriate outcome reporting.^15,17,25 There was significant heterogeneity in the meta-analysis of several recovery events, most likely related to differences in study protocols. In addition, due to the limited number of included studies, it was not possible to perform a funnel plot to test publication bias.

In conclusion, desflurane appears to be associated with a faster recovery from general anaesthesia than sevoflurane in elderly patients. There was no significant between-drug difference in postoperative cognitive function.