Print version ISSN 0034-7094
Rev. Bras. Anestesiol. vol.59 no.2 Campinas Mar./Apr. 2009
Neuroaxis block compared to general anesthesia for revascularization of the lower limbs in the elderly. A systematic review with metanalysis of randomized clinical studies*
Anestesia neuroaxial comparada a la anestesia general para la revascularización de los miembros inferiores en ancianos. Revisión sistemática con metanálisis de ensayos clínicos aleatorios
Fabiano Timbó Barbosa, TSAI; Mário Jorge JucáII; Aldemar Araújo CastroIII
em Ciências da Saúde pela Universidade Federal de Alagoas; Tutor
da Liga de Anestesia, Dor e Terapia Intensiva do Estado de Alagoas
IIPós-Doutorado em Coloproctologia pela Universidade do Texas, Dallas, EUA; Doutor em Gastroenterologia Cirúrgica na Área de Concentração em Coloproctologia; Membro Titular do Conselho e do Colegiado do Curso de Pós-Graduação em Ciências da Saúde do Instituto de Ciências Biológicas e da Saúde da Universidade Federal de Alagoas
IIIMestre em Cirurgia Vascular pela Unifesp; Professor Assistente da Disciplina de Metodologia da Pesquisa Científica; Professor da Universidade de Ciências da Saúde de Alagoas
OBJECTIVES: Currently, it is controversial on whether neuroaxis block (NB)
is more effective than general anesthesia (GA) in elderly individuals undergoing
non-cardiac surgeries. The objective of this study was to determine the efficiency
of NB in comparison to GA for revascularization of the lower limbs (RLL) in
METHODS: A search of the following data base was conducted: MEDLINE (1955 to 2007), CINHAL (1982 to 2007), EMBASE (1980 to 2007), LILACS (1982 to 2007), and ISI (1945 to 2007). Two investigators undertook an independent analysis of the studies published to identify randomized clinical trials (RCTs) comparing NB with GA for RLL. The full text of the RCTs that fulfill the inclusion criteria was analyzed. Disagreements were analyzed in consensus meetings. The software Review Manager was used for the Metanalysis by means of odds ratio with a confidence interval of 95%.
RESULTS: Three studies involving 465 patients were selected. Metanalysis of the following parameters did not show statistically significant differences: mortality (OR: 0.90; CI 95%: 0.30-2.73; p = 0.85 for spinal anesthesia; OR: 1.30, CI 95%: 0.38-4.48, p = 0.68, for epidural block); myocardial infarction (OR: 1.38, CI 95%: 0.29-6.46, p = 0.68); and rate of lower limb amputation (OR: 0.81, CI 95%: 0.30-2.19, p = 0.68, for spinal block; OR: 0.70, CI 95%: 0.24-2.07, p = 0.52 for epidural block). A statistically significant difference was observed for pneumonia (OR: 0.37, CI 95%: 0.15-0.89, p = 0.03); however, clinical heterogeneity was present.
CONCLUSIONS: This metanalysis did not generate enough evidence to demonstrate that NB is more efficient, equivalent, or less efficient than GA for RLL in the elderly.
Key Words: ANESTHETIC TECHNIQUES, General, Regional; SURGERY, Vascular
Y OBJETIVOS: La controversia actual es saber si la anestesia neuroaxial
(AN) es más eficaz que la anestesia general (AG) en ancianos sometidos
a la cirugía no cardíaca. El objetivo fue determinar la eficacia
de la AN comparada con la AG para revascularización de miembros inferiores
(RMI) en ancianos.
MÉTODO: Utilizada la estrategia de búsqueda para las bases de datos: MEDLINE (1955 a 2007), CINHAL (1982 a 2007), EMBASE (1980 a 2007), LILACS (1982 a 2007) y ISI (1945 a 2007). Dos revisores analizaron independientemente los artículos en busca de ensayos clínicos aleatorios (ECA) que comparasen la AN con la AG para la RMI. Se analizó el texto completo de los ECA que respetasen los criterios de inclusión. Las discordancias se analizaron en reuniones consensuales. El metanálisis fue realizado con el software Review Manager, por medio de la razón de chances con intervalo de confianza de un 95%.
RESULTADOS: Se seleccionaron tres artículos originales con 465 pacientes. No hubo significancia estadística en el metanálisis de las variables: mortalidad (OR: 0,90; IC 95%: 0,30 - 2,73; P = 0,85 raquianestesia; OR: 1,30; IC 95%: 0,38 - 4,48; P = 0,68 anestesia epidural), infarto miocárdico (OR: 1,38; IC 95%: 0,29 - 6,46; P = 0,68) y tasa de amputación de los miembros inferiores (OR: 0,81; IC 95%: 0,30 - 2,19; P = 0,68 raquianestesia; OR: 0,70; IC 95%: 0,24 - 2,07; P = 0,52 anestesia epidural). Hubo una significancia estadística para neumonía (OR: 0,37; IC 95%: 0,15 - 0,89; P = 0,03), sin embargo, hubo heterogeneidad clínica.
CONCLUSIONES: Las evidencias generadas en este metanálisis fueron insuficientes para demostrar que la AN es más eficiente, equivalente, o menos eficiente cuando se le compara con la AG para RMI en ancianos.
Developments in surgical and anesthetic techniques decreased perioperative risks and expanded the surgical indications in the elderly 1. In the elderly, general anesthesia has been associated with postoperative respiratory complications 2 and for this reason neuroaxis block has been recommended as the first choice in this population 3.
Regarding elderly patients, and out of the realm of vascular surgery, it has been demonstrated that the mortality associated with neuroaxis block can be 30% lower than that of general anesthesia 3 due to changes inherent to old age, greater number of comorbidities in the elderly, and their greater sensitivity to drugs 1,3; however, a systematic review to determine the efficiency of neuroaxis block in comparison to general anesthesia in elderly patients undergoing revascularization of the lower limbs has not been done.
The present study hypothesized that neuroaxis block is more efficient than general anesthesia for revascularization of the lower limbs in the elderly.
The objective of this study was to determine the efficiency of neuroaxis block, compared to general anesthesia, for revascularization of the lower limbs in the elderly.
This is a systematic review and metanalysis of randomized clinical trials 4. The study was approved by the Ethics on Research Committee of the Universidade Federal de Alagoas. The study was funded by the authors. Original clinical studies, properly randomized, comparing neuroaxis block and general anesthesia for revascularization of the lower limbs were analyzed. Studies were not restricted by the language they were written on. The confidentiality of the allocation was also evaluated.
Adequate randomized distribution was defined as the allocation process that allowed every patient to have the same probability to be allocated to the intervention or experimental group and the control group 5.
A strategy was developed for the search of the following data base: MEDLINE (1955 to 2007), CINHAL (1982 to 2007), EMBASE (1980 to 2007), Latin American and Caribbean Literature on Health Sciences - LILACS (1982 to 2007), and ISI Web of Science (1945 to 2007). The references of the studies that fulfilled the inclusion criteria were also analyzed to identify original studies that were not identified by the search of the data.
Inclusion criteria were as follows: original randomized clinical trials, age above 65 years, and comparing neuroaxis block and general anesthesia. Exclusion criteria included: inadequate randomized distribution, studies with incomplete description, and whenever patients from one of the study groups received combined neuroaxis block and general anesthesia.
Titles and summaries of all studies identified by the search were independently reviewed by two investigators. Afterwards, the full text of this group of studies was requested to evaluate the randomization. Data from studies with adequate randomized distribution were recorded on a standardized form developed by the authors, analyzed, and underwent statistical analysis. The following phases were followed by consensus meetings among the authors.
The Cochrane collaboration's software, Review Manager 6, was used for the statistical analysis. Sensitivity evaluation was planned to explore heterogeneity sources when present. Statistical heterogeneity was quantified by the Chi-square (χ2) and Higgins (I2) tests 7. Odds ratio (OR) with 95% confidence interval (CI) was used to evaluate the statistical difference among the groups analyzed. The random effects model was used.
The primary parameters of the present study included: mortality, myocardial infarction, stroke, muscular paralysis, and postoperative rate of lower limb amputation. Secondary parameters included: length of hospitalization, postoperative cognitive dysfunction, postoperative infection, pneumonia, hema- toma of the neuroaxis, and complications in the recovery room.
The literature search identified 3,913 studies but, based on the inclusion criteria, only 21 were selected. Complete review of those studies showed that three of them, with a total of 465 patients, had adequate randomized distribution and were included in the present study (Table I) 8-10. The references of those three studies, for a total of 158, were also analyzed and two were selected; however, they had already been identified by the search.
The incidence of stroke, muscle paralysis, postoperative cognitive dysfunction, neuroaxis neuroma, complications in the recovery room, degree of satisfaction, postoperative pain, urinary retention, blood transfusion, and length of hospitalization were not analyzed due to the lack of data in the three studies mentioned above.
The spinal block group had a mortality rate of 5%, in the epidural block group it was 4%, and in the general anesthesia group 6% (p = 0.85) and 3% (p = 0.68) when compared to the spinal and epidural blocks, respectively, and statistically significant differences were not observed (Figure 1).
Myocardial infarction had an incidence of 4% in the neuroaxis block group and 3% in the general anesthesia group (p = 0.68), but this difference was not statistically significant (Figure 2).
Lower limb amputation had an incidence of 6% in the spinal block group, 4% in the epidural block group, and 7% in the general anesthesia group (p = 0.68), when compared with the spinal block group, and 6% (p = 0.52) when compared with the epidural block group. Those differences were not statistically significant (Figure 3).
The neuroaxis block group had a 9% incidence of pneumonia, while the general anesthesia group had an incidence of 20% (p = 0.03); the difference was statistically significant and favored the neuroaxis block (Figure 4).
Heterogeneity was not observed in the statistical analyses, but clinical heterogeneity was seen for pneumonia due to the different techniques of neuroaxis block used. One study used spinal block 9 and another used epidural block 9.
The importance of the choice of anesthetic technique for lower limb revascularization has been debated for several years. During this period, clinicians developed strong convictions that one should prefer techniques involving the neuroaxis 11. In the elderly, results from some research centers have demonstrated that postoperative cardiovascular mortality and morbidity of lower limb revascularization do not differ among the different anesthetic techniques used 12; however, a systematic review has not been conducted to corroborate those results in the elderly.
In the present study, 3.913 scientific studies were identified, but only three 8-10 fulfilled the inclusion criteria. Those studies had limitations due to one or more problems, such as: the number of patients was smaller than that predicted by calculating the sample size, difficulties with the double-blind nature of the study in the collection and analysis of the data, exclusions were not described, and the random distribution method and confidentiality were not completely described. The limitations observed in those studies might have influenced their results.
Analysis of the mortality did not show statistically significant differences between both groups (OR: 0.90, CI 95%: 0.30-2.73, p = 0.85 for spinal blocks; OR: 1.30, CI 95%:0.38- 4.48, p = 0.68 for epidural blocks). Other large studies on orthopedic surgery also failed to show statistically significant differences among the different anesthetic techniques in the elderly 13,14. The development of new drugs with a pharmacokinetic profile more suitable to be used in the elderly 1, better understanding of complications 1,2, and the discovery of newer protective strategies 15,16 can justify and even maintain the same results in the next few years.
The incidence of myocardial infarction did not show statistically significant differences between the groups (OR: 1.38, CI95%: 0.29-6.46, p = 0.68). The pathophysiology of perioperative myocardial infarction has not been completely elucidated; however, it seems to be associated with two factors 17: imbalance between myocardial oxygen delivery and consumption, and rupture of the atheromatous plaque with the consequent thrombus formation and coronary obstruction. In anesthesia, this complication can be secondary to the increased oxygen consumption, reduction of oxygen delivery to the myocardium, or both 18. Myocardial infarction can develop in both neuroaxis block and general anesthesia, since the sympathetic blockade seen after the use of local anesthetics can cause a reduction in venous return and left ventricular end-diastolic pressure, and hypotension, resulting in a reduction in coronary perfusion, when it is not properly treated, and general anesthesia might not be able to cause enough reduction in the surgical stress and endocrine response to trauma.
The rate of lower limb amputation did not show statistically significant differences (OR: 0.81, CI 95%: 0.30-2.19, p = 0.68 for spinal blocks; OR: 0.70, CI95%: 0.24-2.07, p = 0.52 for epidural block). The increase in blood flow at the vascular graft is an important determinant factor for the mean life of graft patency 9. The authors of this systematic review assumed the hypothesis that the sympathetic blockade caused by the local anesthetic in the vertebral canal would increase blood flow in the lower limbs, leading to a lower incidence of postoperative amputation; however, the present study did not corroborate this effect. The use of a single dose of local anesthetic was described by the authors of those studies; therefore, one might infer that the vasodilation in the lower limbs was not prolonged, and this could justify the absence of statistically significant differences in the analysis of this parameter, since the authors considered the entire follow-up period and not only the immediate postoperative period.
The development of pneumonia showed statistically significant differences between the groups analyzed (OR: 0.37, CI 95%: 0.15-0.89, p = 0.03). Postoperative pulmonary complications are more common in elderly patients undergoing general anesthesia, not only due to physiological changes characteristic of aging, but also due to the possibility of prolonged mechanical ventilation after the procedure 1. The present study evaluated the incidence of pneumonia, and a greater contribution from one study 8 was observed. In that study 8, the authors report a greater number of elderly individuals and smokers in the general anesthesia group, but they did not describe the rate and correlated their results to the lack of humidification of the anesthetic gases. Analysis of the sensitivity also demonstrated clinical heterogeneity due to different techniques of neuroaxis blockade used by the authors to compare with general anesthesia 8,9. Removal of data from the original assay responsible for the heterogeneity is one of the possible measures to test the results; however, metanalysis can only be done with the data of at least two studies.
We concluded that the evidence generated by this systematic revision with metanalysis was not enough to demonstrate whether neuroaxis block is more efficient, equivalent, or less efficient than general anesthesia in lower limb revascularization in the elderly.
01. Sielenkãmper AW, Booke M - Anaesthesia and the elderly. Curr Opin Anaesthesiol, 2001;14:679-684. [ Links ]
02. Roy RC - Choosing general versus regional anesthesia for the elderly. Anesthesiol Clin North America, 2000;18:91-104. [ Links ]
03. Borgeat A, Ekatdramis G - Orthopaedic surgery in the elderly. Best Pract Res Clin Anaesthesiol 2003;17:235-244. [ Links ]
04. Barbosa FT, Cavalcante JC, Jucá MJ et al. - Neuraxial anaesthesia for lower-limb revascularization (Protocol for a Cochrane Review). Cochrane Database Syst Rev, 2008;(4):CD007083. [ Links ]
05. Altman DG - Randomisation. BMJ, 1991;302:1481-1482. [ Links ]
06. Review Manager (RevMan) [Programa de Computador].Versão 4.2 para Windows. Copenhagen: The Nordic Cochrane Center, The Cochrane Colaboration, 2003. [ Links ]
07. Higgins JP, Thompson SG - Quantifying heterogeneity in a meta-analysis. Stat Med 2002;15:1539-1558. [ Links ]
08. Cook PT, Davies MJ, Cronin KD et al. - A prospective randomized trial comparing spinal anaesthesia using hyperbaric cinchocaine with general anaesthesia for lower limb vascular surgery. Anaesth Intensive Care 1986;14:373-380. [ Links ]
09. Christopherson R, Beattie C, Frank SM et al. - Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery. Anesthesiology 1993;79: 422-434. [ Links ]
10. Pierce ET, Pomposelli Jr FB, Stanley GD et al. - Anesthesia type does not influence early graft patency or limb salvage rates of lower extremity arterial bypass. J Vasc Surg 1997;25:226-232. [ Links ]
11. Tuman KJ, Ivankovich AD - Pro: Regional anesthesia is better than general anesthesia for lower extremity revascularization. J Cardiothorac Vasc Anesth 1994;8:114-117. [ Links ]
12. Yazigi A, Madi-Gebara S, Haddad F et al. - Combined sciatic and femoral nerve blocks for infrainguinal arterial bypass surgery: a case series. J Cardiothorac Vasc Anesth 2005;19:220-221. [ Links ]
13. Gilbert TB, Hawkes WG, Hebel JR et al. - Spinal anesthesia versus general anesthesia for hip fracture repair: a longitudinal observation or 741 elderly patients during 2-yeat follow-up. Am J Orthop 2000;29:25-35. [ Links ]
14. O'Hara DA, Duff A, Berlin JA et al. - The effect of anesthetic technique on postoperative outcomes in hip fracture repair. Anesthesiology, 2000;92:947-957. [ Links ]
15. Dukelgrun M, Schouten O, Feringa HH et al. - Beneficial effects of statins on perioperative cardiovascular outcome. Curr Opin Anaesthesiol, 2006;19:418-422. [ Links ]
16. Bronhein D - Statins and the perioperative period. Semin Cardiothorac Vasc Anesth, 2007;11:231-236. [ Links ]
17. Landesberg G - The pathophysiology of perioperative myocardial infarction: facts and perspectives. J Cardiothorac Vasc Anesth 2003;17: 90-100. [ Links ]
18. Venkataraman R - Vascular surgery critical care: perioperative cardiac optimization to improve survival. Crit Care Med, 2006; 34(9suppl):s200-s207. [ Links ]
Correspondence to: Submitted em 13
de julho de 2008 *
Received from da Universidade Federal de Alagoas, Maceió, AL
Dr. Fabiano Timbó Barbosa
Rua Comendador Palmeira, 113/202 - Farol
57051-150 Maceió, AL
Accepted para publicação em 24 de novembro de 2008
Submitted em 13
de julho de 2008
* Received from da Universidade Federal de Alagoas, Maceió, AL