Age-dependent decrease in minimum alveolar concentration of inhaled anaesthetics: a systematic search of published studies and meta-regression analysis

Editor—A 1996 analysis by Mapleson¹ showed that the minimum alveolar concentration (MAC) of inhaled agents decreases by 6% per decade after age 40 yr. A subsequent analysis by Eger² in 2001 found a similar age-dependent decrease in MAC of 6.7% per decade. An updated determination of the relationship between MAC and age is important to evaluate the reproducibility of prior results when including contemporary studies and to help determine what dose of volatile anaesthetic is appropriate for older patients. To this end, we performed a systematic search for published studies of MAC in individuals of varying ages, and a meta-regression analysis of these data to determine an updated age-dependent decrease in MAC (see Fig. 1).

Fig 1.

Meta-regression of the effect of age on minimum alveolar concentration (MAC) from all published MAC data in individuals of various ages from Table 1, including all those reviewed by Mapleson and all additional MAC studies since 1996.

A systematic literature search was undertaken independently by authors KN and JT, which was then reviewed by MB. All studies in Mapleson's¹ and Eger's² analyses for the agents isoflurane, desflurane, sevoflurane, or halothane were manually searched and included if they met the criteria (discussed below). Further studies were identified with a PubMed database search in March 2018, using the search terms ‘minimum alveolar concentration’ and a specific anaesthetic (isoflurane, sevoflurane, desflurane, or halothane). Study groups were excluded if: the average study group age was <1 yr; the study was not conducted in humans; inhaled anaesthetic was given concomitantly with nitrous oxide, i.v. opioids, or i.v. non-depolarizing neuromuscular blockers; inhaled anaesthetic was given under altered physiologic conditions (hypocapnia, pregnancy, immediate post-partum period); an adjunct drug or treatment (such as i.v. opioids or epidural local anaesthetics) was administered; or subjects had a developmental delay or known central nervous system pathology. No ongoing trials or unpublished studies were included. Data extracted for each study subgroup included: volatile agent given; number of patients; calculated MAC in vol% (mean, standard deviation [sd]/standard error [se]/confidence interval [CI]); age in yr (mean [median when mean value was not reported], sd/se); geographical region (USA, Japan, Europe, India, Canada, China/Hong Kong/Taiwan, Korea, and Israel). Mean MAC and age were collected (or median when mean value was not reported).The meta-regression analysis used similar methods as those used by Mapleson.¹ To estimate change in MAC fraction per decade of age, we constructed a linear meta-regression model of the log-transformed mean MAC on mean age, adjusting for volatile agent administered and geographical region (proxy for altitude), and weighted by (1/number of observations) from each study subgroup. The formulation used was:

$${\text{log}}_{10}\text{MAC}={\beta }_{\text{0}}+{\beta }_1\text{Age}+{\gamma }_i\text{Drug}+{\delta }_j\text{Region}+\epsilon ,$$

where i=1–3 (for halothane, isoflurane, sevoflurane) and j=1–7 (for Japan, Europe, India, Canada, China/Hong Kong/Taiwan, Korea, Israel); desflurane and USA are the reference levels in the model.

We identified 73 patient subgroups from 48 separate studies (Table 1).1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 Of these, 30 patient subgroups had not been included in past meta-analysis studies. There were 1466 patients across all studies, of which 152 (10.4%), 355 (24.2%), 293 (20.0%), and 666 (45.4%) were given desflurane, halothane, isoflurane, or sevoflurane, respectively. The regression analysis showed that MAC decreased by 6.47% (95% CI, 5.17–7.76%) per decade after age 1 yr, with r²=0.98. The volatile anaesthetic agent percentages equal to 1.0 MAC at age 40 yr were 6.44%, 1.16%, and 2.03% for desflurane, isoflurane, and sevoflurane, respectively. The age-dependent decrease slope in MAC was similar for the various anaesthetic agents. In a subset analysis excluding patients over age 75 yr, MAC decreased by 6.59% (95% CI, 5.05–8.10%) per decade, with r²=0.99. Similarly, when excluding patients older than 65 yr, MAC decreased by 6.42% (95% CI, 4.76–8.05%) per decade, with r²=0.99. The similarity between slopes across these meta-regression analyses suggest that the MAC decline slope is log-linear across age, and is not ‘skewed’ by extreme values in patients older than 65 or 75 yr. However, of the 1466 patients included in this meta-regression analysis, only 84 (or 5.7%) were aged ≥65 yr, and only 46 (3.1%) were aged ≥75 yr (Table 1).

Table 1.

Minimum alveolar concentration (MAC) data for desflurane, halothane, isoflurane, and sevoflurane. For studies that were referenced by prior Mapleson¹ and Eger² meta-regressions, we have simply referenced these prior meta-regressions.

Drug	N	Mean age (yr)	MAC (vol%)	First author
Desflurane	12	1.8	8.72	Eger2
	12	3.7	8.62	Eger2
	12	7.5	7.98	Eger2
	9	25.6	7.25	Eger2
	15	27	6	Wadwha5
	15	33	6.2	Wadwha5
	11	48	6	Eger2
	24	52	6.54	Lin6
	24	52	6.7	Lin6
	18	69.6	5.17	Eger2
Halothane	12	35.2	0.76	Eger2
	16	41.7	0.74	Eger2
	8	30	0.75	Eger2
	24	42.1	0.76	Mapleson1
	10	40.6	0.81	Drasner7
	9	31	0.74	Fahey8
	18	39.1	0.73	Eger q2
	12	10.6	0.9	Frei9
	14	1.5	0.97	Eger2
	19	4.1	0.91	Eger2
	8	8.4	0.87	Eger2
	11	15.5	0.92	Eger2
	22	24.9	0.84	Eger2
	24	42	0.76	Eger2
	10	81.4	0.64	Eger2
	12	28.6	0.7	Houghton10
	10	35.2	0.68	Houghton10
	18	37.5	0.7	Houghton10
	12	38	0.77	Inagaki11
	14	25	0.65	Kalman12
	11	31.2	0.78	Licina13
	16	20.6	0.73	Eger2
	11	1.2	0.94	Eger2
	24	42	0.77	Eger2
	10	58	0.78	Schwieger14
Isoflurane	16	44.1	0.85	Aantaa15
	12	2	1.6	Eger2
	12	4	1.6	Eger2
	20	5.4	1.5	Chen16
	10	34.5	1.075	Gin17
	20	38	1.28	Johansen18
	24	34.91	1.14	Kashyap19
	10	6.1	1.69	Liu20
	72	55	1	Eger2
	11	1.4	1.69	Eger2
	16	26.1	1.28	Eger2
	32	44.2	1.15	Eger2
	30	64	1.05	Eger2
	8	27.4	1.04	Zhou21
Sevoflurane	93	49.6	0.94	Byon22
	89	49.9	0.93	Byon22
	36	36	2.14	Ezri23
	27	40	2.34	Ezri23
	39	41	1.91	Ezri23
	15	62.9	1.77	Eger2
	12	76	1.45	Eger2
	13	25	2.6	Eger2
	30	51	1.86	Hamp24
	20	5	2.22	Eger2
	22	39	1.95	Eger2
	20	47.6	1.71	Eger2
	20	4.31	2.49	Eger2
	42	48.7	1.58	Eger2
	12	1.7	2.6	Eger2
	12	3.9	2.5	Eger2
	12	7.7	2.5	Eger2
	18	41	2.2	Luo25
	24	82	1.22	Eger2
	20	71.4	1.48	Eger2
	16	38	2.05	Eger2
	27	44	1.9	Eger2
	21	41.5	2.2	Wajima26
	26	42	2.2	Wajima27

We found a high degree of reproducibility and similarity between the age-dependent decrease values for MAC found in this study and two prior studies.^1,2 This is reassuring both for our scientific understanding of how anaesthetic sensitivity changes with age, and for our clinical understanding of how these drugs should be dosed in older patients. This study has four main limitations. First, including data from prior meta-regressions^1,2 may have reduced potential differences from previous estimates of age-dependent changes in MAC. Second, few studies have measured MAC in patients older than 75 yr, so estimates of age-dependent changes in MAC over age 75 yr remain largely an extrapolation. Third, we used subgroup summary measures from the cited studies as we did not have access to individual subject data, which could possibly underestimate variability in the raw data. Fourth, we measured MAC, not MAC-Awake. However, Eger's² analysis notes that the ratio of MAC-Awake to MAC for isoflurane and sevoflurane remains the same as age increases. This suggests that the age-dependent decrease in MAC-Awake is likely similar to the age-dependent decrease in MAC determined here. Overall, this updated measurement of a 6.47% decrease in MAC per decade of age can help anaesthesia providers determine appropriate end-tidal anaesthetic concentrations in older patients, and can serve as a standard benchmark for quantitatively evaluating the extent to which anaesthesia providers are appropriately adjusting inhaled anaesthetic dose for patient age.^3,4

Authors' contributions

Study design: all authors.

Study conceptualization: MB.

Literature search: KN, JT.

Literature search adjudication: MB.

Data interpretation: all authors.

Data analysis: MC.

Drafting of the manuscript: KN, MC, MB.

Editing of the manuscript: JT, DG, TH, TM, MJ, MK, MB.

All authors made substantial contributions to the conception and design, acquisition of data, or analysis and interpretation of data for this study; drafted or revised the article critically for important intellectual content; gave their final approval of the version to be published; and agree to be accountable for all aspects of the work thereby ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declarations of interest

MB acknowledges income from a legal consulting case related to postoperative cognition in an older adult, and material support (i.e. EEG monitors) from Massimo for a study unrelated to the data presented here. The other authors have no conflicts to disclose.

Funding

Foundation for Anesthesia Education and Research Medical Student Anesthesia Research Fellowship (to KN). NIH grants R03-AG050918 and K76-AG057022 (to MB), and a Jahnigen Scholar Award from the Foundation for Anesthesia Education and Research and the American Geriatrics Society (to MB). MB also acknowledges additional support from NIH grants T32-GM08600 and P30AG028716, and the Duke Anesthesiology Department.