PaCO2 Six months after the Initiation of Long-term Noninvasive Ventilation in Patients with COPD

Transcription

1 ORIGINAL ARTICLE PaCO2 Six months after the Initiation of Long-term Noninvasive Ventilation in Patients with COPD Tomomasa Tsuboi 1,3, Toru Oga 1, Kazuko Machida 2,KensukeSumi 3, Susumu Oguri 3, Atsuo Sato 3, Takuya Kurasawa 3, Motoharu Ohi 4, Michiaki Mishima 5 and Kazuo Chin 1 Abstract Background and Objective The appropriate target level for PaCO2 after the introduction of long-term noninvasive positive pressure ventilation (NPPV) in patients with COPD remains uncertain, and therefore must be tested. Methods Data on 54 patients with COPD receiving long-term domiciliary NPPV were examined retrospectively. PaCO2 a few months after NPPV and potential confounders were analyzed with discontinuation of long-term NPPV as the primary outcome. The differences in annual hospitalization rates due to respiratory deterioration between those from 1 year before to 2 years after initiation of NPPV were compared according to the PaCO2 measured at 6 months after NPPV (6-mo PaCO2). Results 6-mo PaCO2 seemed to be most related to continuation of NPPV (p=0.019). Patients with 6-mo PaCO2 of less than 60 mmhg had maintained a significantly lower PaCO2 value 6 to 24 months after NPPV (p=0.04) and had a significantly higher continuation rate of NPPV (p=0.03) than those with a 6-mo PaCO2 of 60 mmhg or more. Annual hospitalization rates due to respiratory deterioration were not associated with the 6-mo PaCO2 level, but fatal hospitalization rates during the first year of NPPV were significantly correlated with relatively high 6-mo PaCO2 (p=0.008). Conclusion A relatively low 6-mo PaCO2 value was predictive of long-term use of NPPV. The target values of 6-mo PaCO2 may, therefore, be less than 60 mmhg in COPD patients with extremely severe hypercapnia, although more prospective studies are needed. Key words: chronic respiratory failure, home mechanical ventilation, noninvasive positive pressure ventilation, partial pressure of carbon dioxide, chronic obstructive pulmonary disease (Intern Med 50: , 2011) () Introduction Non-invasive positive pressure ventilation (NPPV) has been widely used in the treatment of patients with chronic hypercapnic respiratory failure (1-4), while the evidence in support of long-term NPPV in COPD is equivocal (5, 6). NPPV has been shown to improve gas exchange, probably due to increased ventilatory response to carbon dioxide, in patients with COPD (1, 2, 5-10). A significant decrease in PaCO2 during the first and second year following the institution of NPPV was reported in COPD patients (3). At present, however, the target level for PaCO2 after introduction of long-term domiciliary NPPV in patients with COPD remains uncertain. The general condition of patients and arterial blood gas levels (ABGs) are unstable during the pre-nppv period even when NPPV is started during a chronic state. Gas exchange and clinical status, however, markedly stabilize after institution of long-term NPPV. Therefore, for the prognostic factors, the parameters after a few months of NPPV seem to be more appropriate than those before the start of NPPV. Department of Respiratory Care and Sleep Control Medicine, Kyoto University Graduate School of Medicine, Japan, Department of Respiratory Medicine, National Tokyo Hospital, Japan, Department of Respiratory Medicine, National Hospital Organization Minami-Kyoto Hospital, Japan, Sleep Medical Center, Osaka Kaisei Hospital, Japan and Department of Respiratory Medicine, Kyoto University Hospital, Japan Received for publication August 2, 2010; Accepted for publication December 2, 2010 Correspondence to Dr. Tomomasa Tsuboi, tsuboit@skyoto.hosp.go.jp 563

2 NPPV aimed at maximal reduction of PaCO2 has been reported to improve ABGs and achieve a high survival rate in patients with stable hypercapnic COPD (8, 11, 12). Those studies, however, did not assess the association of PaCO2 levels a few months after NPPV with the success of longterm NPPV. Therefore, to verify that COPD patients with relatively low PaCO2 after NPPV have a better prognosis than those with higher values, we retrospectively examined the association of PaCO2 a few months after NPPV on continuation rates of NPPV and on hospitalization rates due to respiratory deterioration. Also, we wanted to clarify which levels of PaCO2 a few months after NPPV were suitable for patients with COPD. Patients Methods Fifty-four COPD patients who had started long-term domiciliary NPPV at 6 hospitals affiliated with Kyoto University Hospital and the National Tokyo Hospital from November 20, 1991 to May 2, 2007, were included in this retrospective study. Diagnosis of COPD was made according to the American Thoracic Society guidelines (13). All patients had chronic respiratory failure with hypercapnia. The NPPV therapy was begun either after an acute episode or during the chronic state. The decision for initiation of longterm NPPV was based on clinical symptoms with persistent hypercapnia during daytime (PaCO2>55 mmhg) and/or nocturnal hypoventilation and/or clinical instability with recurrent hospitalizations. Patients with other causes of chronic respiratory failure such as post-tuberculosis, kyphoscoliosis, neuromuscular disorders, obesity hypoventilation syndrome, or bronchiectasis were excluded. The patients received NPPV nocturnally or received it for approximately two hours in the daytime in addition to night-time NPPV. Of the 54 patients, 38 patients were divided into 2 groups according to PaCO2 measured at 6 months after NPPV (6- mo PaCO2) (Group-1, <60 mmhg, n=23; Group-2, 60 mmhg, n=15). Measurement of 6-mo PaCO2 could not be performed in the remaining patients because of clinical instability at the time or because they did not visit the outpatient clinic. The 54 patients were followed up until November 30, Clinical surveys had been performed at the end of every year from 1995 to 2002, in December 2004 and in December Measurements Age at the start of NPPV, gender, BMI, FEV1%, presence of sleep disorder suggesting sleep apnea assessed by patients symptoms and nocturnal pulse oxymetry (4% oxygen desaturation index more than 5), presence of comorbidities (Charlson comorbidity index) (14), status upon introduction of NPPV (i.e., acute or chronic state), duration of long-term oxygen therapy (LTOT) before the start of NPPV, ventilator mode (assisted or controlled) and other ventilator settings, inspiratory positive airway pressure (IPAP), expiratory positive airway pressure (EPAP), and backup respiratory rate (fr) were all examined and/or documented for risk factors. Annual number of hospitalizations due to respiratory deterioration (acute bronchitis, pneumonia, spontaneous pneumothorax, chronic disease progression, aggravation of cor pulmonale, etc.), and annual number of hospitalizations for fatal exacerbations from 1 to 2 years after the start of NPPV were also examined. Daytime ABGs were examined from 12 months before the start of NPPV to the observable end-point if available. Determinations of ABGs were made with the patient in the supine position breathing prescribed oxygen without NPPV support. ABGs were obtained in patients who were in a stable condition without exacerbation except for those obtained at the start of NPPV from patients administered NPPV during an acute state. All data about the clinical course of patients were collected from clinical records. Clinical protocol for introducing long-term NPPV The precise criteria and procedure for introducing longterm NPPV have been described elsewhere (15). At the start of NPPV, volume preset ventilators were utilized with custom fabricated nasal masks (16) or with commercially available interfaces. Pressure preset ventilators using bilevel PAP devices were applied with commercially available interfaces. Supplemental oxygen was added to NPPV to maintain SaO2 >95% during diurnal NPPV and >90% during nocturnal NPPV. This study was approved by the Ethics Committee of Kyoto University, and individual consent was not obtained as stipulated by the Committee. According to the recommendation of the Ethics Committee of Kyoto University, the protocol of this study is posted on the website of our institute and all inquiries are answered. Follow up In Japan, patients receiving long-term NPPV and/or LTOT must visit an outpatient clinic each month. ABG analysis and chest X-ray were usually performed every 3 to 6 months. Statistical analysis Means and SDs are reported for patients characteristics at the start of NPPV and for ventilator settings. Unpaired t tests for continuous variables and chi-square tests for categorical variables were used to compare differences in patients characteristics between the 2 groups (Group -1 and - 2) of patients divided by 6-mo PaCO2 levels. Effects on continuation rates of long-term NPPV of either PaCO2 or bicarbonate (HCO3 - ) measured at 12 months before to 12 months after initiation of NPPV were examined using univariate Cox proportional hazards regression analyses. Comparisons of continuation rates of long-term NPPV by several risk factors including 6-mo PaCO2 were performed using univariate and multivariate Cox proportional hazards regression analy- 564

3 Table 1. Patient Characteristics at Start of NPPV (n=50) ses. Continuation rates of long-term NPPV were also assessed using Kaplan-Meier analyses (log-rank test). Between Groups -1 and -2, comparisons of annual hospitalization rates and annual fatal hospitalization rates were performed using unpaired t tests. To evaluate differences of PaCO2 over time between the 2 groups, repeated measures analysis of variance was used. Patient characteristics Results A total of 54 COPD patients who had continued NPPV for more than two months were followed. Since four patients discontinued domiciliary NPPV due to improvement in hypercapnic respiratory failure, analyses were undertaken in 50 patients. Of these 50 patients, 46 used pressure preset ventilators and 4 used volume preset ventilators. Table 1 shows characteristics of the entire study population (n=50) and the subgroups formed according to the 6- mo PaCO2 (Group -1 and Group -2). Overall, the patients were characterized by severe obstructive ventilatory defects, severe hypercapnia, severe malnutrition and an unstable clinical condition. At the start of long-term NPPV, FEV1% was lower and PaCO2 was higher in Group -2 patients. In the univariate analysis, among ABG measurements at 12 months before to 12 months after initiation of NPPV, lower PaCO2 and lower HCO3 - measured at both 6 and 12 months after initiation of NPPV were all associated with higher continuation rates of NPPV (Table 2). Overall outcome The outcome in 50 patients was as follows. After the long-term use of NPPV, 2 Group -1 patients switched to long-term tracheostomy positive pressure ventilation (TPPV) and died four months and 37 months after the switch, respectively. No Group -2 patient nor any ungrouped patient changed to long-term TPPV. By December 2007, 23 patients (11/23 Group -1 patients; 12/15 Group -2 patients; 0/12 ungrouped patients) had died and 27 continued NPPV. Of the 23 deaths, 16 (8 Group -1; 8 Group -2) were due to progression of respiratory failure, 2 (1 Group -1; 1 Group -2) from lung cancer, 2 (2, Group -2) from colon cancer, 2 (2, Group -1) from sepsis due to ileus, and 1 (1, Group -2) from an unknown cause (sudden death at home). Time course of PaCO2 before and after NPPV The comparison of the time course of PaCO2 between 565

4 Table 2. Effects on Continuation Rates of Long-term NPPV of Either PaCO 2 or Bicarbonate (HCO 3 -) Measured at 12 Months before to 12 Months after Initiation of NPPV (Univariate Modality Model). Figure 1. Time course of PaCO 2 before and after NPPV. Comparison between patients divided by average PaCO 2 6 months after introduction of long-term NPPV (6-mo PaCO 2) (Group-1, <60 mmhg; Group-2, 60 mmhg). Data are presented as mean (SD). From 6 to 24 months after the initiation of NPPV, patients with a relatively high 6-mo PaCO 2 had a significantly higher PaCO 2 than those with a relatively low 6-mo PaCO 2 (p=0.04). Groups 1 and 2 is shown in Fig. 1. From 6 to 24 months after the initiation of NPPV, Group -1 patients had a significantly lower PaCO2 than Group -2 patients (p=0.04). Comparison of rates of continuation of long-term NPPV In the univariate analysis, 6-mo PaCO2 were significantly associated with higher continuation rates of NPPV (p<0.01), 566

5 Table 3. Comparisons of Continuation Rates of Long-term NPPV by Several Risk Factors Including 6-mo PaCO 2 (Univariate Modality Model). Table 4. Comparisons of Continuation Rates of Long-term NPPV by Several Risk Factors (Multivariate Modality Model). while a higher BMI, higher FEV1%, and the presence of a sleep disorder suggesting sleep apnea tended to be associated with higher continuation rates (p<0.1) (Table 3). In the multivariate analysis that included BMI, FEV1%, presence of sleep disorder suggesting sleep apnea and 6-mo PaCO2, only 6-mo PaCO2 was significantly associated with continuation rates of NPPV (Table 4). Results of the Kaplan-Meier analysis showed significantly increased continuation rates of long-term NPPV in Group-1 patients (p= 0.005; Fig. 2). The 2- and 5-year probabilities of continuing NPPV in Group -1 were 86.7% and 54.9%, respectively, and those for Group -2 were 53.3% and 17.8%, respectively. Comparison of annual hospitalization rates between two groups divided by 6-mo PaCO2 There was no significant difference in annual hospitalization rates due to respiratory deterioration between Group -1 and Group -2, from the year preceding NPPV to the 2nd year of long-term NPPV. However, Group -2 patients had significantly higher hospitalization rates for fatal exacerbations of the 1st year of NPPV (p=0.008) (Fig. 3). Discussion In the present study, patients with a relatively low PaCO2 6 months after initiation of NPPV had significantly higher continuation rates of NPPV and significantly lower hospitalization rates for fatal exacerbations in the first year of NPPV than those with higher PaCO2 values. Such patients maintained a relatively low PaCO2 from 6 to 24 months after initiation of NPPV. The role of long-term NPPV for treatment of COPD patients with hypercapnic respiratory failure is still controversial (5, 6). While short-term studies showed that domiciliary NPPV improved physiological parameters and quality of life (1, 2, 7, 8, 17), no significant reduction in mortality through the use of NPPV was reported in two randomized controlled trials (RCT) (18, 19). Recently, however, an RCT 567

6 Figure 2. Kaplan-Meier curves of continuation rates of long-term NPPV. Comparison between patients divided by average PaCO 2 at 6 months after the introduction of long-term NPPV (6-mo PaCO 2) (Group -1, <60 mmhg; Group -2, 60 mmhg). Patients with a relatively low 6-mo PaCO 2 had a significantly better prognosis (log-rank test, p=0.005). Figure 3. Comparison of hospitalization rates in each year among patients divided by PaCO 2 at 6 months after introduction of long-term NPPV (6-mo PaCO 2) (Group -1, <60 mmhg; Group -2, 60 mmhg). Data are presented as mean (SD). Patients with a relatively low 6-mo PaCO 2 had significantly lower hospitalization rates for fatal exacerbations in the first year of NPPV (p=0.008). showed that nocturnal NPPV in stable oxygen-dependent patients with COPD may improve survival (20). In Japan, long-term NPPV was estimated to be prescribed for approximately 4,500 patients with COPD in 2004, which represented about 30% of all NPPV users (21). COPD was also reported to become one of the major indications for home 568

7 mechanical ventilation in Europe (22). Those reports imply that through clinical experience physicians have gained strong confidence in the beneficial effects of long-term NPPV for some COPD patients with severe hypercapnia. In general, patients with hypercapnic respiratory failure might not be completely stable. Among the present patients, PaCO2 increased slowly during the pre-nppv period and ABGs in cases starting NPPV in an acute setting were further disturbed. It has been reported that daytime PaCO2 values just before NPPV could not predict the continuation rate of NPPV (12). Therefore, ABGs before NPPV seem inappropriate as predictive variables. Gas exchange and the clinical state, however, markedly stabilized after a few months of NPPV in patients who started long-term NPPV in either a chronic or acute state. Thus, parameters a few months after the introduction of NPPV seem to be more valuable as predictive factors than pre-nppv parameters with regard to all patients starting NPPV who thereafter live under the same effective treatment. In the other words, it is supposed that a relatively short-term (a few months) effect of domiciliary NPPV should impact its relatively long-term (a few years) effect. A lower base excess (BE) at the start of NPPV and a larger decrease in BE after the introduction of NPPV were shown to be associated with a better prognosis while a lower daytime PaCO2 at the start of NPPV was not (11). BE may reflect the long-term respiratory response to chronic persistent hypercapnia. Unfortunately, compared with PaCO2, BE is a less familiar parameter to general chest physicians. In the present study, as shown in Table 2, 6-mo PaCO2 could predict continuation rates of NPPV to a degree similar to HCO3 - as a substitute for BE 6 months after initiation of NPPV. Recently, NPPV using relatively high IPAP in a controlled (timed) mode has been demonstrated as feasible in patients with hypercapnic COPD (17). We also reported that, in patients with restrictive thoracic disease (RTD), a pure controlled mode resulted in significantly higher continuation rates of long-term NPPV than an assisted mode (15). However, in the present study, there was no difference in continuation rates of NPPV between the use of those two ventilator modes in COPD patients. Since COPD patients with chronic hypercapnic respiratory failure were not reported to have low-frequency fatigue of the diaphragm (23), there is a possibility that in COPD patients merely passive ventilation and respiratory muscle rest induced by NPPV with a controlled mode might be insufficient to improve patients prognosis. Long-term NPPV is not as effective in patients with COPD as in those with RTD (5). Therefore, a larger decrease in night-time PaCO2 during NPPV leading to reduced daytime PaCO2 might be essential for an increase in continuation rates of long-term NPPV. Elliott emphasized that hypercapnia is a poor prognostic sign in COPD, and that more aggressive ventilation might result in a larger decrease in PaCO2 (9). A higher level of ventilation can be obtained if greater support pressure is used (24). Normalization of PaCO2 by NPPV in patients with hypercapnic respiratory failure due to COPD is possible and leads to a significant reduction of PaCO2 during spontaneous breathing (7). Moreover, NPPV using relatively high inspiratory pressures has been shown to improve ABGs and survival rates in patients with stable hypercapnic COPD (8, 11, 12). In the present study, patients with a relatively high 6-mo PaCO2 had more severe ventilatory defects at the start of NPPV. While long-term NPPV was shown to decrease the PaCO2 level in both groups (Groups -1 and -2), relatively high 6-mo PaCO2 values persisted in Group -2, that is, in those with the higher PaCO2 levels, throughout their clinical course. These results indicate the possibility that patients with more advanced disease tend to have higher PaCO2 than those with less advanced disease, with or without the application of long-term NPPV. Therefore, it remains unclear whether further lowering of PaCO2 in each patient is possible. Beginning and prevalence of NPPV were delayed for more than five years in Japan compared to European countries (22). Therefore, in many COPD patients with hypercapnic respiratory failure, LTOT had been the only applicable treatment and their PaCO2 had already become elevated to extremely high levels before the start of long-term NPPV. Thus, in some of our patients, daytime PaCO2 could not be reduced below 60 mmhg with long-term NPPV although high IPAP levels were used. For such patients, the start of long-term domiciliary NPPV might be too late. Anyway, as we took into account the feasibility and the comfort of NPPV, we did not insistently try to reduce PaCO2 with high intensity ventilatory support in patients whose daytime PaCO2 remained greater than 60 mmhg. Efforts to reduce the daytime PaCO2 to a lower level, if possible, might improve the prognosis. Trials of high intensity NPPV for COPD patients with extremely severe hypercapnia appear to be necessary. We recognize that our study had several limitations. First, the results need to be verified by large-scale prospective studies. Secondly, we have not provided validated and objective data on the physiological effects of NPPV on gas exchange and respiratory muscles. In conclusion, we first found that patients with a relatively low 6-mo PaCO2 maintained such levels for considerable periods and that 6-mo PaCO2 was a significant predictive factor of rates of continuing NPPV and hospitalization for fatal exacerbations in the first year of NPPV. Therefore, the temporary target level of daytime PaCO2 after NPPV may be less than 60 mmhg, although some more prospective studies are needed to clarify whether reducing daytime PaCO2 below 60 mmhg is possible and/or beneficial for individual patients with extremely severe hypercapnia and to elucidate to what extent daytime PaCO2 should be reduced in individual patients with moderate to severe hypercapnia. The authors state that they have no Conflict of Interest (COI). 569

8 Acknowledgement We thank Hitoshi Murao, Seisuke Niibayashi, Naoki Sakai, Kenichi Takahashi, Yoshiko Kawabe, Kazue Shimada, Kazuo Endo, Yuichi Chihara, Kensaku Aihara, Kiminobu Tanizawa, and Kunihiko Kamakari for their contribution to this study. References 1. Elliott MW, Mulvey DA, Moxham J, Green M, Branthwaite MA. Domiciliary nocturnal nasal intermittent positive pressure ventilation in COPD: mechanisms underlying changes in arterial blood gas tensions. Eur Respir J 4: , Meecham Jones DJ, Paul EA, Jones PW, Wedzicha JA. Nasal pressure support ventilation plus oxygen compared with oxygen therapy alone in hypercapnic COPD. Am J Respir Crit Care Med 152: , Leger P, Bedicam JM, Cornette A, et al. Nasal intermittent positive pressure ventilation: long-term follow-up in patients with severe chronic respiratory insufficiency. Chest 105: , Simonds AK, Elliott MW. Outcome of domiciliary nasal intermittent positive pressure ventilation in restrictive and obstructive disorders. Thorax 50: , Goldberg A, Leger P, Hill N, et al. Clinical indication for noninvasive positive pressure ventilation in chronic respiratory failure due to restrictive lung disease, COPD, and nocturnal hypoventilation - a consensus conference report. Chest 116: , Kolodziej MA, Jensen L, Rowe B, Sin D. Systematic review of noninvasive positive pressure ventilation in severe stable COPD. EurRespirJ30: , Windisch W, Vogel M, Sorichter S, et al. Normocapnia during nippv in chronic hypercapnic COPD reduces subsequent spontaneous PaCO2. Respir Med 96: , Windisch W, Kosti S, Dreher M, Virchow Jr JC, Sorichter S. Outcome of patients with stable COPD receiving controlled noninvasive positive pressure ventilation aimed at a maximal reduction of PaCO2. Chest128: , Elliott MW. Noninvasive ventilation in chronic ventilatory failure due to chronic obstructive pulmonary disease. Eur Respir J 20: , Nickol AH, Hart N, Hopkinson NS, et al. Mechanisms of improvement of respiratory failure in patients with COPD treated with NIV. Int J Chro Obstruct Pulmon Dis 3: , Windisch W, Haenel M, Storre JH, Dreher M. High-intensity noninvasive positive pressure ventilation for stable hypercapnic COPD. Int J Med Sci 6: 72-76, Budweiser S, Jorres RA, Riedl T, et al. Predictors of survival in COPD patients with chronic hypercapnic respiratory failure receiving noninvasive home ventilation. Chest 131: , American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 136: , Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40: , Tsuboi T, Oga T, Machida K, et al. Importance of ventilator mode in long-term noninvasive positive pressure ventilation. Respir Med 103: , Tsuboi T, Ohi M, Kita H, et al. The efficacy of a custom fabricated nasal mask on gas exchange during NIPPV. Eur Respir J 13: , Dellweg D, Schonhofer B, Haidl PM, et al. Short-term effect of controlled instead of assisted noninvasive ventilation in chronic respiratory failure due to chronic obstructive pulmonary disease. Respir Care 52: , Casanova C, Celli BR, Tost L, et al. Long-term controlled trial of nocturnal nasal positive pressure ventilation in patients with severe COPD. Chest 118: , Clini E, Sturani C, Rossi A, et al. The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients. Eur Respir J 20: , McEvoy RD, Pierce RJ, Hillman D, et al. Nocturnal non-invasive nasal ventilation in stable hypercapnic COPD: a randomized controlled trial. Thorax 64: , Ishihara H. Ventilatory assist therapy. Nippon Rinsho 65: , 2007 (in Japanese, abstract in English). 22. Lloyd-Owen SJ, Donaldson GC, Ambrosino N, et al. Patterns of home mechanical ventilation use in Europe: results from the Eurovent survey. Eur Respir J 25: , Schonhofer B, Polkey MI, Suchi S, Kohler D. Effect of home mechanical ventilation on inspiratory muscle strength in COPD. Chest 130: , Tuggey JM, Elliott MW. Titration of non-invasive positive pressure ventilation in chronic respiratory failure. Respir Med 100: , The Japanese Society of Internal Medicine 570