Anticholinergic premedication for flexible bronchoscopy a randomized

Document Sample
Anticholinergic premedication for flexible bronchoscopy a randomized Powered By Docstoc
					                                                                                   Nicholas Vozoris
                                                                         Journal Club February 2010
                    Anticholinergic premedication for flexible bronchoscopy:
        a randomized, double-blind, placebo-controlled study of atropine and glycopyrrolate
                             Malik JA et al., Chest 2009; 136: 347-354

       Anticholinergics are used around the world as premedication for bronchoscopy, with the
       proposed benefits being improved visualization of the tracheobronchial tree through (1)
       antisecretory effects; (2) prevention of bronchoconstriction.
       However, the evidence in support of anticholinergic premedication for bronchoscopy is slim, with
       previous studies based on small numbers of subjects and not adjusting for potential confounders.
       The purpose of this study was to examine the utility and safety of two anticholinergic drugs,
       atropine and glycopyrrolate, in an unselected cohort of patients referred for bronchoscopy.

      Randomized, double-blind, placebo-controlled trial
      Subjects: 1,000 consecutive patients undergoing elective bronchoscopy at a tertiary care hospital
      in India between January 2003 – May 2004.
      Inclusion criteria: all inpatients & outpatients > 15 years old undergoing elective bronchoscopy
      Exclusion criteria: previous participation in similar study, undergoing brachytherapy or
      therapeutic bronchoscopy, history of glaucoma or prostate disorders, baseline hypoxemia <90%,
      using supplemental oxygen, intubated patients.
      Lab assistant not involved with bronchoscopy randomly allocated patients (although no details
      given on randomization process) to one of 3 premedication regimens 20-40 minutes pre-
           o Atropine 0.01mg/kg IM
           o Glycopyrrolate 0.005mg/kg IM
           o Placebo (20cc saline solution) IM
      Bronchoscopist & patient blinded to premedication…but MD may be able to speculate on type of
      premedication given based on degree of secretions, HR/BP response, etc.
      All received midazolam 0.07mg/kg IV pre-procedure & subsequent doses as needed for anxiety
      and agitation, plus nebulized lignocaine 4%. Amounts of extra midazolam & lignocaine given not
      specified across groups.
      Bronchoscopies were attempted initially transnasally, with oral approach used if significant nasal
      blockage present (95% transnasally). Biopsies were performed as needed.
           1) bronchoscopist & patient feedback on amount of airway secretions, cough, and patient
               discomfort, using a 100mm visual analogue scale (VAS), with labels of worst on the
               extreme left at 0mm, and most favourable on the extreme right at 100mm. Respondents
               asked to mark the point on the line that best described their observation. Distance of this
               point measured from left line origin and expressed as mm (although looking at tables,
               they appear to express VAS scores in the opposite direction, which causes confusion).
               This item is presumably the primary outcome, although not clearly specified.
           2) willingness of patient to undergo repeat bronchoscopy.
           3) oxygen desaturation via finger oximetry
           4) bradycardia/tachycardia
           5) hypotension
           6) procedure time
      No power calculation described.

   Proportion of smokers different across groups – smokers may be more likely to have secretions &
   obstructive airways disease. Smoking status is controlled though for in logistic regression models.
   Some relevant items that are ‘missing’ from in Table 1 include mean midazolam & lignocaine
   doses used during bronchoscopy, other medication use (e.g., inhalers), baseline respiratory
   symptoms, and relevant comorbidities like asthma and COPD.
    Critical appraisal
        Randomization? Yes, although no details given about randomization process.
        Follow-up complete? Yes
        Intention-to-treat? Presumably yes
        Double-blinding? Technically yes, but MD may have been able to speculate on type of
        premedication given based on degree of secretions, HR/BP response, etc.
        Groups similar? Proportion of smokers is different across groups, and other relevant items listed
        on previous page not included in Table 1….hard to say for sure if groups were truly similar.
        Groups treated equally? Same point as above.
        What were the results?
            o Presentation of VAS score results confusing; they appear to present VAS scores in the
                 opposite direction from what was outlined in methods section.
            o While the differences in bronchoscopist VAS scores for airway secretions are statistically
                 significant across groups, there are unlikely to be of any clinical significance (e.g., is
                 there a significant clinical difference in a score of 31mm vs 26 mm on a VAS?).
            o There are also no improvements in other outcomes measured (i.e., cough, patient
                 discomfort, patient willingness to return for repeat procedure, oxygen saturation, total
                 procedure time), to further corroborate the premise that anticholinergic premedication is
            o Although stastically significant, the potential harms of anticholinergics not very
                 impressive either – is a several more point rise in HR or sBP with anticholergics
                 compared to placebo clinically significant?
        Can the results be applied to my patients? To some degree yes, as outpatients with a wide
        range respiratory problems included. However, subjects mostly men, there is no information on
        obstructive airway diseases, and no inpatients are included.
        Clinically important outcomes considered? Yes, degree of secretions and cough are important
        outcomes, but the way these outcomes were measured is somewhat problematic (i.e., VAS
        scores). Why not measure volume of secretions suctioned back, or amount of extra midazolam or
        lidocaine needed to be given, instead?
        Treatment/benefits worth potential cost/harm? Not designed to assess cost-effectiveness.

My conclusions
Well-intentioned study to help make the procedure of bronchoscopy more evidence-based, but the few
‘positive’ results of anticholinergic premedication are likely of no clinical significance. At the same time,
anticholinergic medication does not appear to be harmful…overall anticholinergics just appear useless.