Journal of Clinical Anesthesia
Volume 24, Issue 1 , Pages 3-7, February 2012

Meta-analysis of the effect of central neuraxial regional anesthesia compared with general anesthesia on postoperative natural killer T lymphocyte function☆

  • Ian Conrick-Martin, FCAI, MRCPI, DIBICM (Specialist Registrar in Anaesthesia)

      Affiliations

    • Department of Anaesthesia, Mater Misericordiae University Hospital, Eccles St., Dublin 7, Ireland
    • Corresponding Author InformationCorrespondence: Ian Conrick-Martin, FCAI, MRCPI, DIBICM, Department of Anaesthesia, Mater Misericordiae University Hospital, Eccles St., Dublin 7, Ireland. Tel.: +353 1 803-2281; fax: +353 1 830-9563.
    • ,
  • Malcolm R. Kell, MD, FRCSI, FRCS(Glasg) (Consultant Surgeon)

      Affiliations

    • Department of Surgery, Mater Misericordiae University Hospital, Eccles St., Dublin 7, Ireland
    • ,
  • Donal J. Buggy, MD, MSc, FRCPI, FCAI, FRCA (Professor of Anaesthesia; Consultant in Anaesthesia)

      Affiliations

    • Department of Anaesthesia, Mater Misericordiae University Hospital, Eccles St., Dublin 7, Ireland
    • Outcomes Research Consortium, Cleveland Clinic, OH, USA

Received 4 September 2010; received in revised form 3 August 2011; accepted 6 September 2011.

Article Outline

Abstract 

Study Objective

To compare the effect of central neuraxial (spinal or epidural) anesthesia with general anesthesia on postoperative natural killer (NK) T lymphocyte function.

Design

Meta-analysis.

Setting

University-affiliated hospital.

Measurements

A systematic search of the medical literature from 1966 to 2009 yielded 5 eligible studies with a total of 184 patients who received neuraxial blockade. Natural killer T lymphocyte function was studied.

Main Results

There was significant heterogeneity between the studies [I2 = 94.4% (95% CI= 90.3-96.2%)]. Overall fixed-effect odds ratio was 0.86 (0.66-1.14, P = 0.25). The random-effect odds ratio was 1.13 (0.26-4.92, P = 0.79).

Conclusion

Anesthetic technique does not appear to significantly affect postoperative NK T lymphocyte function. Given the heterogeneity observed, further clinical studies in cancer patients of the effect of anesthetic technique on immune function in general, and NK T lymphocyte function in particular, are needed.

Keywords: Anesthesia: general, regional neuraxial, Cancer patients, Natural killer T lymphocyte, Surgical stress response

 

Back to Article Outline

1. Introduction 

There is emerging interest in evaluating the hypothesis that anesthetic technique during primary cancer surgery influences the incidence of cancer recurrence or metastasis [1], [2], [3]. Indeed, a randomized clinical trial is underway to investigate this hypothesis [4]. Surgery and the stress response to surgical tissue injury are associated with transient perioperative inhibition of immune function in general, and natural killer (NK) T lymphocytes in particular [5], [6]. These are large granular lymphocytes, the majority of which express CD16 and CD56 cell surface antigens, which induce lysis of abnormal (eg, infected or tumor) target cells [7]. Experimental study in a live animal model of injected cancer cells has shown that surgery per se facilitates cancer metastases and simultaneously inhibits NK T lymphocyte function [8].

Certain anesthetic agents and techniques, particularly regional anesthesia, may attenuate the metastatic spread in this experimental model and preserve NK T lymphocyte function [9-11]. However, it is unclear whether clinical studies of human perioperative immune function have shown an advantage of regional anesthetic techniques over general anesthesia in terms of NK T lymphocyte function.

Therefore, our hypothesis was that central neuraxial regional anesthesia preserved perioperative NK T lymphocyte function relative to general anesthesia. We undertook a meta-analysis of all clinical studies investigating the effect of central neuraxial (spinal or epidural) regional anesthesia compared with general anesthesia on perioperative NK T lymphocyte function.

Back to Article Outline

2. Materials & methods 

We conducted an electronic search of the literature in the National Library of Medicine’s PubMed (MEDLINE 1966 – April 2009) database and EMBASE (between 1980 - April 2009). Electronic search terms included “regional an(a)esthesia”, “natural killer lymphocyte (cell) function”, “immune function”, “T-lymphocyte function”, “lymphocyte activity”, “stress response”, “general an(a)esthesia”, “epidural an(a)esthesia”, “spinal an(a)esthesia”, and “intrathecal an(a)esthesia”. This search was supplemented by hand-searching of relevant anesthesia, immunology, and general medical journals. Further information was sought from scanning reference lists of already retrieved papers. The initial data abstraction was performed independently by one of the authors (ICM) and the resulting data-set was verified by consensus between all authors. Any disagreements were resolved by discussion between the authors.

2.1. Eligibility criteria 

Only reports of patients undergoing surgery with epidural or spinal anesthesia as a study arm and NK T lymphocyte function as an endpoint were included. We included trials regardless of the exact type of anesthesia, as long as this was standardized to the treatment arm; however, the primary endpoint of this meta-analysis was alteration in NK T lymphocyte function.

We included all human adult studies published in English in peer-reviewed literature in which NK T lymphocyte function was measured in the context of use of central neuraxial anesthesia. We initially identified 394 potentially relevant studies, of which 389 were subsequently excluded from meta-analysis. The trial flow chart is shown in Fig. 1. We did not include abstracts, editorials, or studies in which the primary outcome variables could not be measured. No duplicate population was included in the meta-analysis.

The main reasons for exclusion from analysis included the lack of NK T lymphocyte function as an endpoint measurement, the non-use of central neuraxial anesthesia as a study arm, and the lack of analyzable data in the study and/or control groups. Studies were reviewed by the authors and agreement was reached regarding eligibility, with disagreements regarding the correct categorization of data being resolved by consensus decision. Four of the 5 studies randomized patients to the study arms; one did not (Koltun et al [12]). Given the impractical nature of blinding patients, clinicians, or investigators to a control group without a neuraxial intervention and an “experimental” group with such an intervention, blinding was not utilized in any of the studies.

2.2. Data extraction and outcomes 

We recorded the following information regarding each eligible trial: authors' names, journal, year of publication, study design items (including whether there was a description of the mode of randomization, allocation concealment, number of withdrawals per arm, and blinding), and the number of patients assigned to treatment and analyzed per arm. The primary outcome was change in NK T lymphocyte function after surgery. Following a review of the literature, it was evident that no consistent outcome measurement for NK T lymphocyte function exists. The outcomes analyzed in the studies used in this meta-analysis included NK T lymphocyte absolute numbers, NK lymphocyte percentage, NK cell cytotoxicity, and NK T lymphocyte activity. NK T lymphocyte activity is also an expression for cytotoxicity measured in vitro in a Cr-release cytotoxicity assay in peripheral blood mononuclear cells. Individual study outcome measurements were expressed as a percentage change in each group (ie, study and control) and subsequently as a percentage difference between the two groups. Figures were corrected to two decimal places.

2.3. Statistical analysis 

Data were extracted by manual retrieval from the available tables, figures, and text of the individual articles. If further information was required, it was requested directly by contacting the authors concerned. We calculated individual values for each study data point using percentages and estimated specific values from the overall (n) value. Validation of the data extraction was confirmed by consensus amongst the authors.

For NK T lymphocyte function in each study, we estimated the odds ratio (OR), with its variance and 95% confidence interval (CI) between epidural or spinal anesthesia, and general anesthesia. In studies that did not provide explicit accounting of the numbers of events for each arm, we derived the pertinent data from information available in the published reports. Heterogeneity between the ORs for the same outcome between different studies was assessed by chi-square-based Q statistic as well as the I2 statistic used to describe statistical heterogeneity [9].

Data were then combined across studies by general variance methods with fixed-effects and random-effects models. Fixed-effects analysis weighted the natural logarithm of each study's risk ratio by the inverse of its variance. Random-effects analysis weighted the natural logarithm of each study's risk ratio by the inverse of its variance plus an estimate of the between-study variance in the presence of between-study heterogeneity. In the absence of between-study heterogeneity, fixed and random effects coincide because the between-study variance is zero [13]. Analyses were conducted using Statsdirect version 2.5.6 (StatsDirect, Ltd, Cheshire, UK) and SPSS version 12.0 (SPSS, Inc., Chicago, IL, USA). All statistical tests were two-tailed.

Back to Article Outline

3. Results 

Of the 394 studies reviewed, 5 met the inclusion criteria, with a total of 184 patients [[12], [14]-17]. Table 1 lists the characteristics of these studies. The only methods of regional anesthesia used in trials in which NK T-lymphocytes were studied were those involving central neuraxial anaesthesia (ie, epidural and spinal anesthesia).

Table 1. Individual study details
Authors/year of publicationAnesthetic techniqueType of surgeryIndication for surgeryMean age (yrs)Study populationGender (m/f)Outcome measured
Koltun et al, 1996 [12]Epidural vs opioidOpen colectomyVaried54 vs 69203/7 vs 7/3NKCC
Tønnesen and Wahlgreen, 1988 [14]Epidural vs opioidAbdominal hysterectomyMenorrhagia (nonmalig)41 vs 45160/8 vs 0/8NKCC
Yokota et al, 2000 [15]Intrathecal opioid vs IV opioid vs controlHysterectomyUterine myoma43 vs 44 vs 434020 vs 10 vs 10NK activity
Volk et al, 2004 [16]Epidural vs opioidLumbar spinal surgeryUnknown42 vs 515412/15 vs 7/20NK cell number
Ahlers et al, 2008 [17]Intraop epidural vs opioid & postop epiduralUpper/lower GI resectionUnknown55 vs 595416/11 vs 18/9NK cell %
184 (total)

NKCC = natural killer cytotoxicity, nonmalig = nonmalignant, NK = natural killer, IV = intravenous, Intraop = intraoperative, postop = postoperative, GI = gastrointestinal.

3.1. Types of surgeries 

The types of surgeries patients underwent were varied: Koltun et al involved colectomies; Tønnesen et al and Yokota et al both dealt with hysterectomies; Volk et al involved lumbar spinal surgery; and Ahlers et al studied upper and lower gastrointestinal resections.

3.2. Underlying diagnosis 

The underlying diagnoses studied (eg, oncological or other condition) were also varied. Koltun et al included a variety of diagnoses. Tønnesen et al involved nonmalignant causes of menorrhagia. Yokota et al studied uterine myomata. The diagnoses in the cohorts enrolled in the studies by Volk et al and Ahlers et al were unclear in both papers.

3.3. Neuraxial blockade 

Koltun et al involved thoracic epidurals that were continued until the third postoperative day. Tønnesen et al studied lumbar epidurals which were continued up to the fourth postoperative day. Yokota et al studied spinal anesthesia with variable doses of intrathecal morphine. Volk et al included lumbar epidurals which were continued until the fourth postoperative day. Ahlers et al involved thoracic epidurals, the duration of which was unclear.

3.4. Measurement of NK T lymphocyte function 

Koltun at al included the collection of blood specimens on the day prior to surgery, within 24 hours after surgery, and on postoperative day three. Tønnesen et al involved the collection of blood specimens preoperatively, intraoperatively, and until the fifth postoperative day. Yokota collected blood specimens preoperatively, two hours after surgery, and on the first and second postoperative days. Volk et al involved the collection of blood specimens preoperatively, immediately after surgery, and on the first, third, and seventh postoperative days. Ahlers et al involved the collection of blood specimens preoperatively, two hours after commencement of surgery, and on postoperative days one and four. The methodology of measurement of NK T lymphocyte function is discussed above.

Given the relatively small number of studies involved, epidural and spinal anesthesia groups were not analyzed separately. There was a significant degree of heterogeneity between the studies (Cochran Q value = 70.9, P < 0.0001), with an I2 value of 94.4% (95% CI = 90.3-96.2%). The fixed-effects and random-effects OR meta-analysis plots are shown in Fig. 2. The overall fixed-effect OR was 0.86 (95% CI = 0.66-1.14, P = 0.25) and the random-effect OR was 1.13 (95% CI= 0.26-4.92, P = 0.79).

  • View full-size image.
  • Fig. 2. 

    Fixed-effects and random-effects odds ratio meta-analysis plots. The overall fixed-effect odds ratio (OR) was 0.86 (95% CI = 0.66-1.14, P = 0.25) and the random-effect OR was 1.13 (95% CI = 0.26 - 4.92, P = 0.79).

Back to Article Outline

4. Discussion 

NK T lymphocytes form part of the innate immune response. There is considerable evidence that anesthesia and major surgery produce significant perioperative immunosuppression in humans, with profound alterations detectable in cell-mediated immunity [18]. In the context of these perioperative immunological changes, NK T lymphocytes play a crucial role in influencing tumor development, particularly in the metastatic process. Suppression of NK T lymphocyte activity occurs rapidly after surgery and is aggravated by the degree of surgical damage to tissues, pain, anesthetic and analgesic medications, and psychological stress [5-11].

Despite experimental evidence linking regional anesthesia with preservation of NK T lymphocyte function and resistance of cancer metastases in a live animal model, we have shown that there are few clinical studies evaluating the effect of anesthetic technique on immunological function, and NK T lymphocyte function in particular. This meta-analysis of 5 studies showed no advantage of central neuraxial regional anesthesia over general anesthesia in NK T lymphocyte function.

In addition to surgery per se, general anesthesia is immunosuppressive and tumor-promoting; many drugs used to provide anesthesia suppress NK T lymphocyte activity and promote metastasis [10]. Concurrently, the administration of opioids seems also to be immunosuppressive with inhibition of components of cellular and humoral immune function such as antibody production, NK lymphocyte activity, cytokine secretion, lymphocyte proliferative response to mitogens, and phagocytic activity [11], [15].

These data should be interpreted with caution because they are limited by heterogeneity among the studies included, in terms of the type of surgery that patients underwent, whether they had pre-existing cancer or not, and the disparate nature of various techniques of evaluating T lymphocyte function described. For example, NK T lymphocyte function as a study endpoint may be reported in terms of NK T lymphocyte numbers, NK lymphocyte percentages, NK cell cytotoxicity, or NK T lymphocyte activity. Due to differences between these methods of measurement, we used percentage changes for the purposes of this meta-analysis. Compilation of actual data from the various studies would have been preferable in terms of reliability of summating the data but this was not possible in this exercise. Opioids also influence NK T cell function [11], [15], and these were used in combination with local anesthetics for some of the neuraxial techniques reported, making interpretation of overall results uncertain. As with all meta-analyses, there exists the possibility of publication bias whereby trials with positive findings are preferentially accepted for publication. This situation was limited by the methodical nature of our review of the literature.

Therefore, while the conclusion of this present meta-analysis is that central neuraxial anesthetic techniques appear to have no overall effect in preserving NK T lymphocyte function perioperatively, further clinical studies clearly are warranted in patients undergoing primary cancer surgery. Such studies should include all known methods of evaluating NK T lymphocyte function and a composite assessment of the aggregate effect of these tests.

Back to Article Outline

References 

  1. Ó Riain SC, Buggy DJ, Kerin MJ, Watson RW, Moriarty DC. Inhibition of the stress response to breast cancer surgery by regional anesthesia and analgesia does not affect vascular endothelial growth factor and prostaglandin E2. Anesth Analg. 2005;100:244–249
  2. Exadaktylos AK, Buggy DJ, Moriarty DC, Mascha E, Sessler DI. Can anesthetic technique for primary breast cancer surgery affect recurrence or metastasis?. Anesthesiology. 2006;105:660–664
  3. Biki B, Mascha E, Moriarty DC, Fitzpatrick JM, Sessler DI, Buggy DJ. Anesthetic technique for radical prostatectomy surgery affects cancer recurrence: a retrospective analysis. Anesthesiology. 2008;109:180–187
  4. Sessler DI, Ben-Eliyahu S, Mascha EJ, Parat MO, Buggy DJ. Can regional analgesia reduce the risk of recurrence after breast cancer? Methodology of a multicenter randomized trial. Contemp Clin Trials. 2008;29:517–526
  5. Galley HF, Dubbels AM, Webster NR. The effect of midazolam and propofol on interleukin-8 from human polymorphonuclear leukocytes. Anesth Analg. 1998;86:1289–1293
  6. Buggy DJ, Smith G. Epidural anaesthesia and analgesia: better outcome after major surgery? Growing evidence suggests so. BMJ. 1999;319(7209):530–531
  7. Trinchieri G. Biology of natural killer cells. Adv Immunol. 1989;47:187–375
  8. Ben-Eliyahu S, Page GG, Yirmiya R, Shakhar G. Evidence that stress and surgical interventions promote tumor development by suppressing natural killer cell activity. Int J Cancer. 1999;80:880–888
  9. Bar-Yosef S, Melamed R, Page GG, Shakhar G, Shakhar K, Ben-Eliyahu S. Attenuation of the tumor-promoting effect of surgery by spinal blockade in rats. Anesthesiology. 2001;94:1066–1073
  10. Melamed R, Bar-Yosef S, Shakhar G, Shakhar K, Ben-Eliyahu S. Suppression of natural killer cell activity and promotion of tumor metastasis by ketamine, thiopental, and halothane, but not by propofol: mediating mechanisms and prophylactic measures. Anesth Analg. 2003;97:1331–1339
  11. Shavit Y, Ben-Eliyahu S, Zeidel A, Beilin B. Effects of fentanyl on natural killer cell activity and on resistance to tumour metastasis in rats. Dose and timing study. Neuroimmunomodulation. 2004;11:255–260
  12. Koltun WA, Bloomer MM, Tilberg AF, et al. Awake epidural anesthesia is associated with improved natural killer cell cytotoxicity and a reduced stress response. Am J Surg. 1996;171:68–72
  13. Lau J, Ioannidis JP, Schmid CH. Quantitative synthesis in systematic reviews. Ann Intern Med. 1997;127:820–826
  14. Tønnesen E, Wahlgreen C. Influence of extradural and general anaesthesia on natural killer cell activity and lymphocyte subpopulations in patients undergoing hysterectomy. Br J Anaesth. 1988;60:500–507
  15. Yokota T, Uehara K, Nomoto Y. Intrathecal morphine suppresses NK cell activity following abdominal surgery. Can J Anaesth. 2000;47:303–308
  16. Volk T, Schenk M, Voigt K, Tohtz S, Putzier M, Kox W. Postoperative epidural anesthesia preserves lymphocyte, but not monocyte, immune function after major surgery. Anesth Analg. 2004;98:1086–1092
  17. Ahlers O, Nachtigall I, Lenze J, et al. Intraoperative thoracic epidural anaesthesia attenuates stress-induced immunosuppression in patients undergoing major abdominal surgery. Br J Anaesth. 2008;101:781–787
  18. Pirtikkangas CO, Salo M, Riutta A, Perttilä J, Peltola O, Kirvela O. Effects of propofol and Intralipid on immune response and prostaglandin E2 production. Anaesthesia. 1995;50:317–321

☆ Supported by the Sisk Foundation, Dublin, and the Eccles Breast Cancer Research Fund, Dublin, Ireland.

PII: S0952-8180(11)00326-6

doi:10.1016/j.jclinane.2011.09.001

Journal of Clinical Anesthesia
Volume 24, Issue 1 , Pages 3-7, February 2012

Article Tools

* Image Usage & Resolution