Effectiveness of acetazolamide for reversal metabolic alkalosis

in weaning COPD patients from mechanical ventilation

Christophe Faisy, Amel Mokline, Olivier Sanchez, Jean-Marc Tadié, Jean-Yves Fagon

ELECTRONIC SUPPLEMENTARY MATERIAL

DETAILS ON SUBJECTS AND METHODS

Definition of case-patients

COPD was diagnosed according to the criteria of the American Thoracic Society [1]. To take into account the presence of mixed acid-base balance disturbance (chronic respiratory acidosis and metabolic alkalosis) in COPD patients, metabolic alkalosis was defined by serum bicarbonate >26 mmol/l and pH ≥7.38 daily measured by arterial blood gas analysis (ABL 725Ò, Radiometer Copenhagen, Copenhagen, Denmark) performed on pressure support ventilation or volume-assisted ventilation (Evita 4Ò, Dräger Medical France, Antony, France) [2-4]. Weaning period was defined as the time between the readiness to wean and extubation. Criteria for assessing readiness to wean were the following: resolution of disease acute phase for which the patient was intubated; no sedation and Coma Glasgow Score >12; absence of excessive tracheobronchial secretions and adequate cough; body temperature ≤38.5°C; hemodynamic stability (no vasopressors, heart rate <120 beats/min, systolic blood pressure 90-160 mmHg); stable metabolic status (no significant abnormalities in plasma electrolytes and serum glucose levels); hemoglobin level >8 g/dl; respiratory frequency ≤35 breaths/min; positive end-expiratory pressure support (PEEP) ≤8 cmH2O; no significant respiratory acidosis; fractional inspired of oxygen (FiO2) ≤50% and PaO2/FiO2 ≥150 mmHg. Patients with an intolerance or allergy to ACET or other sulfonamide-based agents, who had received ACET or sodium bicarbonate in the previous 72 h, who had vomiting or diarrhea or hypokalemia (serum potassium concentration <3.6 mmol/l), who had non-compensated acidosis (arterial pH <7.38) or serum creatinine of >200 μmol/l or who received dialysis, were not included in the study. Patients with previous study inclusion were also excluded. Thirty consecutive COPD patients were eligible for enrolment during the study period. Four patients were excluded: two had acute renal failure and two had non compensated acidosis with arterial pH <7.38.

Definition of controls

Patients who fulfilled the criteria for COPD, admitted for acute respiratory failure but with no acetazolamide (ACET) treatment during their ICU stay were eligible. The inclusion criteria for the control group were: metabolic alkalosis as defined above, COPD males and females between 45 and 90 years of age, forced expiratory volume in 1 s (FEV1) and FEV1/forced vital capacity (FVC) <70% predicted, current smokers or ex-smokers with at least 20 pack-years, Simplified Acute Physiology Score II (SAPS II) [5] at ICU admission between 25 and 75. The patients of the control group were matched in a proportion 1:1 with each case-patient on the variables of metabolic alkalosis at readiness to wean, age, and severity of illness at ICU admission. Ranges for matching values were constructed a priori within reasonable nearness: ± 4 mmol/l for serum bicarbonate; ± 0.05 for arterial pH; ± 5 years for age; ± 5 for SAPS II. The exclusion criteria were the same ones as the case-patients group. Patients who experienced unplanned extubation (self-extubation or endotracheal malfunction) were also excluded. Sixty five COPD patients were screened and 26 controls (40%) were retained for matching.

Weaning from mechanical ventilation

Weaning process was standardized according to the recommendations of the Consensus Conference of the Société de Réanimation de Langue Française held in October 2001. Weaning strategy was progressive decrease in pressure support ventilation or volume-assisted ventilation with the use of progressively increased time on a T-piece and was chosen by physicians in charge according to the difficulty of the weaning process [6]. External PEEP was applied to counterbalance intrinsic PEEP, typically 4-6 cmH2O, in COPD patients. During weaning process, salbutamol, dobutamine or furosemide were administered intravenously when indicated by attending physicians. Spacer and nebulizer for administration of salbutamol via the respiratory circuit were not used in our setting because of the requirement of ventilator patterns adjustment to optimize the inhaled mass of drug and of the unpredictable fraction of drug deposited in the lower respiratory tract. Corticosteroid therapy was not used. Caloric intakes were calculated with the Faisy 2003 equation and were administered via a nasogastric tube [7].

Criteria for extubation and reintubation

In our ICU, criteria for extubation and reintubation were also standardized in a written protocol. Patients were extubated if they tolerated at least 1 h of spontaneous breathing trial via a T-piece oxygenated with the same FiO2 employed during mechanical ventilation. Arterial blood gas analysis was performed at the end of the T-piece trial. Criteria for passing spontaneous breathing trial included: respiratory frequency ≤35 breaths/min; decrease in arterial oxygen saturation <5% or PaO2/FiO2 ≥150 mmHg; increase in PaCO2 <10 mmHg and decrease in pH <0.10; stable hemodynamic status (increase in heart rate or systolic blood pressure <20%); stable mental status; no need for frequent suctioning of airway secretions. If patients satisfied the spontaneous breathing trial criteria, they were extubated in presence of a physiotherapist and received supplemental oxygen to provide arterial oxygen saturation ≥90%. Weaning was considered successful if reintubation or non-invasive ventilation was not required within 48 h after extubation. Criteria for considering reintubation or non-invasive ventilation were: agitation or depressed mental status; tachypnoea ≥35 breaths/min; tachycardia, hypertension, hypotension or arrhythmia; inability to clear pulmonary secretions or presence of upper airway obstruction; increase in PaCO2 ³10 mmHg and decrease in pH ³0.10; PaO2 <60 mmHg or arterial oxygen saturation <90% while receiving FiO2 >50%.

Data collection

Data collected included age, sex, smoking, use of long-term systemic corticosteroids or diuretics, home oxygen therapy or home non-invasive ventilation, body mass index at ICU admission, left ventricular ejection fraction (LVEF) assessed by echocardiography and prior pulmonary function tests performed in stable condition, cause of respiratory failure, SAPS II at ICU admission, and arterial blood gas at the time of intubation. The variations (D) in arterial blood gas and respiratory parameters were assessed between study inclusion (baseline) and extubation. To assess the respiratory acidotic defect of metabolic alkalosis occurring in compensation of prolonged respiratory acidosis, the respiratory acid and the pure metabolic components of alkalosis were calculated at baseline and at extubation as follows: assuming a normal arterial blood gas of pH 7.40, serum bicarbonate 24 mmol/l, and a rise in 3.5 mmol/l of bicarbonate serum level for every 10 mmHg increase of PaCO2 for chronic respiratory acidosis [8, 9], respiratory acidotic component (mmol/l) = (PaCO2 – 40) × 0.35 and pure metabolic component (mmol/l) = serum bicarbonate – (24 + respiratory acidotic component). All daytime arterial blood gas and respiratory parameters were collected at 7.00 am on pressure support or volume-assisted ventilation. Serum levels of electrolytes involved in acid-base equilibrium (sodium, potassium, and chloride) and the haemoglobin levels (involved in CO2 transport) were also assessed at 7.00 am. ACET tolerance was checked every day. The respiratory parameters were measured using the spirometry module of the Evita 4Ò ventilator which was calibrated prior to each use. We also collected duration of weaning (from readiness to wean till the first planed extubation), number of failed spontaneous breathing trials before extubation, success rate of weaning, length of invasive mechanical ventilation, length of ICU stay, occurrence of nosocomial pneumonia, and ICU outcome. The diagnosis of nosocomial pneumonia was based on the following criteria: systemic signs of infection (fever, tachycardia, and leukocytosis), new or worsening infiltrates seen on the chest roentgenogram, and bacteriologic evidence of pulmonary parenchymal infection (protected specimen brush with >103 or bronchoalveolar lavage with >104 colony-forming units/ml) [10].

Statistical analysis

Results are presented as numbers (%), means ± SD or as medians (range) for data with non normal distribution. The statistics were calculated with StatView 4.5 Software (Abacus Concepts Inc, Berkeley, CA). The sample size was calculated a priori based on the effect of ACET on serum bicarbonate decrease in hypercapnic patients with COPD: a minimal difference of 6 ± 5.3 mmol/l was expected [4, 11]. To detect such difference (we used a b risk of 10% and an a error of 5%), ³52 patients (26/group) were required. Continuous variables were analyzed using the Student’s t-test for normally distributed variables and the Mann-Whitney test for non-normally distributed data. The categorical variables were analyzed using the Chi-square test with Yates corrections or Fisher’s exact test when necessary. The correlation coefficient r was calculated by using linear regression analysis. A p value < 0.05 was considered to be statistically significant.

DETAILS ON RESULTS

Characteristics of the patients

Descriptive and demographic characteristics of the 26 patients included and their 26 matched-controls are summarized in Table 1. At the time of intubation, ACET patients and their controls had comparable arterial blood gas and the proportion of patients with bicarbonate >26 mmol/l was similar between groups (Table 1). At readiness to wean, PaCO2 and pH were significantly improved (all p values < 0.05 vs. intubation) whereas PaO2/FiO2 ratio and bicarbonate did not differ significantly from values collected at the time of intubation (all p values > 0.05) in the two groups (Tables 1, 2). Comparing those who received and those who did not receive ACET, there were no significant differences in terms of modalities of ventilation or baseline respiratory data during weaning procedure (Table 2). ACET patients and their controls had similar haemoglobin concentration, serum sodium and chloride except serum potassium at the time of extubation (Table 2). The use of salbutamol (7.6% vs. 11.5%), dobutamine (3.8% vs. 7.6%) or furosemide (34.6% vs. 46.1%) in the weaning period did not differ significantly between ACET patients and their matched-controls (all p values > 0.05). ACET patients and matched-controls were similar for the median duration of furosemide treatment [2 (1-4) vs.1.5 (1-6) days, respectively, p = 0.54] and for the cumulative dose of furosemide (76 ± 26 vs. 97 ± 30 mg, respectively, p = 0.61) administered during weaning period.

Effects of ACET on arterial blood gas

In ACET patients, serum bicarbonate level at readiness to wean strongly correlated with the drop of bicarbonate induced by ACET administration (r = 0.75, p < 0.0001). Only 9/26 (34.6%) ACET patients temporary achieved a complete reversal of their metabolic alkalosis during weaning. The respiratory acidotic component of alkalosis remained similar between ACET patients and their controls through the weaning period (Fig. 1). In contrast, ACET patients had a higher pure metabolic component of alkalosis at baseline and this component was significantly reduced by ACET treatment during weaning procedure (Fig. 1).

Patient outcome

ACET patients and their matched-controls did not differ significantly for the length of mechanical ventilation [8.5 (2-32) vs. 12.5 (2-43) days, respectively, p = 0.09], ICU stay [15 (6-120) vs. 19 (4-73) days, respectively, p = 0.21], and ICU mortality [6 (23%) vs. 5 (19.5%), respectively, p = 0.73].


ADD ON DISCUSSION

The present study shows that ACET, used at a single daily dose of 500 mg, reverses incompletely metabolic alkalosis and fails to change PaCO2 and respiratory parameters in weaning COPD patients from invasive mechanical ventilation. Advanced age and high severity of disease on ICU admission explain for the most part the prolonged acute mechanical ventilation and the high rate of extubation failure in our COPD patients [12, 13], maybe contributing to ACET failure. However, inefficiency of ACET should be also interpreted in regard to the changes in acid-base and respiratory parameters, and to the dose regimen used in our study.

The effects of ACET on the ventilatory response of COPD patients are complex and not easily predictable compared to its effects in healthy individuals [4, 14, 15] or in experimental models [16, 17]. Metabolic acidosis generated by ACET, acting on the central and peripheral chemoreceptors, is the strongest factor to increase minute ventilation, and the ventilatory response generally reduces the arterial carbon dioxide tension by 5 to 6 mmHg [14]. In non-intubated COPD, Vos and colleagues reported that ACET reduced serum bicarbonate and dropped pH and PaCO2 [18]. In our study, we observed that ACET-induced change in serum bicarbonate remained moderate (around 3.4 mmol/l) without effects on carbon dioxide tension or minute ventilation. We also found the respiratory acidotic component of alkalosis remained similar between ACET patients and their controls while ACET was only effective for reducing the pure metabolic component of alkalosis. In non-intubated COPD patients, the physiologic compensatory metabolic response is an adequate reaction occurring in patients with chronic CO2 retention. Conversely, the pure alkalotic component depends on extra-respiratory situations causing metabolic alkalosis and is likely to depress the central respiratory drive. Conflicting with these findings, we observed that the ACET-induced decrease in pure metabolic component of alkalosis had no impact on minute ventilation and PaCO2 in weaning COPD patients. Indeed, if artificial ventilation is used to reduce the PaCO2, the patients are left with excess serum bicarbonate and, therefore, with metabolic alkalosis. In our study, ACET patients and their controls had similar arterial blood gas at the time of intubation and at readiness to wean, and the proportion of patients with serum bicarbonate >26 mmol/l was comparable between groups. Finally, in COPD patients on mechanical ventilation, the pure metabolic component of alkalosis appears as the result of a complex interaction of multiple factors including reventilation and caution is needed to interpret its changes under treatment with ACET. Further physiological studies are needed to clarify the complex interaction between the components of mixed metabolic alkalosis, the activity of respiratory central drive and mechanical ventilation in COPD patients.

Caution is needed in extrapolating our results beyond our patient recruitment and standard procedures of care. Indeed, the COPD patients enrolled in our study experienced respiratory failure from various causes requiring drugs interfering with acid-base balance and altering ACET effectiveness. However, our patient recruitment reflects the variety of clinical situations needing invasive mechanical ventilation in decompensated COPD (“the true life”) and warrants, at least in part, the external validity of our study. Moreover, case-control studies are subject to bias in selecting cases and controls, cannot study variables that may be altered by the disease event, and may present recall and bias problems in measuring exposure [19].

References

  1. American Thoracic Society (1995) Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD). Am J Resp Crit Care Med 152:S77– S120
  2. Webster NR, Kulkarni V (1999) Metabolic alkalosis in the critically ill. Crit Rev Lab Sci 36:497–510
  3. Khanna A, Kurtzman A (2001) Metabolic alkalosis. Respir Care 46:354–365
  4. Mazur JE, Devlin JW, Peters MJ, Jankowski MA, Iannuzzi MC, Zarowitz BJ (1999) Single versus multiple dose of acetazolamide for metabolic alkalosis in critically ill patients: a randomized double-blind trial. Crit Care Med 27:1257–1261
  5. Le Gall JR, Lemeshow S, Saulnier F (1993) A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 270:2957–2963

6.  Boles JM, Bion J, Connors A, Herridge M, Marsh B, Melot C, Pearl R, Silverman H, Stanchina M, Vieillard-Baron A, Welte T (2007) Weaning from mechanical ventilation. Eur Respir J 29:1033–1056