The Obesity Hypoventilation Syndrome Can Be Treated With Noninvasive Mechanical Ventilation

The Obesity Hypoventilation Syndrome Can Be Treated With Noninvasive Mechanical Ventilation* Juan F. Masa, MD; Bartolome R. Celli, MD, FCCP; Juan A. Riesco, MD; Manuel Herna´ndez, MD; Julio Sa´nchez de Cos, MD; and Carlos Disdier, MD

Study objectives: To assess the effectiveness of nasal noninvasive mechanical ventilation (NIMV) in patients with obesity hypoventilation syndrome (OHS). Design: Clinical assay that compares two groups of patients with hypercapnic respiratory failure, one group with OHS and the other group with kyphoscoliosis, in their basal situation and after 4 months of treatment with nocturnal NIMV. Thirty-six patients (22 patients with OHS and 14 patients with kyphoscoliosis) completed the study protocol. Results: The frequency of symptoms, such as morning headache, morning drowsiness, dyspnea, and leg edema, improved in a statistically significant way in both groups of patients. The sleepiness improved only in the group with OHS. The comparison of frequency of symptoms between both groups of patients after NIMV treatment did not present statistically significant differences. In the resting situation and without nasal ventilation in place, the PO2 (mean ⴞ SD) changed from 51 ⴞ 10 to 64 ⴞ 11 mm Hg (p < 0.001) and PCO2 from 58 ⴞ 10 to 45 ⴞ 5 mm Hg (p < 0.001) when the patients with OHS were treated with NIMV. In the group of patients with kyphoscoliosis, likewise without nasal ventilation in place, PO2 changed from 53 ⴞ 6 to 65 ⴞ 5 mm Hg (p < 0.001) and PCO2 from 59 ⴞ 11 to 45 ⴞ 4 mm Hg (p < 0.001) with NIMV treatment. When we compared PO2 and PCO2 in both groups of patients at the beginning and at the end of NIMV treatment, we did not find statistically significant differences between OHS and kyphoscoliosis. Conclusions: NIMV improves the clinical symptoms and the respiratory failure of patients with OHS to a similar degree to that reported for diseases in which its use is completely established, such as kyphoscoliosis. Therefore, NIMV could be an alternative to the treatment of patients with OHS. (CHEST 2001; 119:1102–1107) Key words: home mechanical ventilation; hypercapnic respiratory failure; kyphoscoliosis; noninvasive mechanical ventilation; obesity; obesity hypoventilation syndrome Abbreviations: AHI ⫽ apnea/hypopnea index; CPAP ⫽ continuous positive airway pressure; NIMV ⫽ noninvasive mechanical ventilation; OHS ⫽ obesity hypoventilation syndrome

obesity hypoventilation syndrome (OHS) is T hecharacterized by the coexistence of obesity and hypercapnic respiratory failure. It is sometimes associated with obstructive sleep apnea. The origin of this form of respiratory failure is not clear. A central origin associated with mechanical alterations of the lung or chest because of obesity has been proposed.1 Noninvasive mechanical ventilation (NIMV) has been successfully used in chest wall diseases (mainly *From the Pulmonary Division (Drs. Masa, Riesco, Herna´ndez, Sa´nchez de Cos, and Disdier), “San Pedro de Alca´ntara” Hospital, Ca´ceres, Spain; and St. Elizabeth’s Medical Center (Dr. Celli), Boston, MA. This work was performed at “San Pedro de Alca´ntara” Hospital, Ca´ceres, Spain. Manuscript received January 31, 2000; revision accepted October 19, 2000. Correspondence to: Juan F. Masa, MD, C/Rafael Alberti 12, 10001 Ca´ceres, Spain; e-mail: [email protected] 1102

kyphoscoliosis),2 in central alveolar hypoventilation syndrome,3 and also in nighttime hypoventilation of obese patients without daytime respiratory failure.4 However, no studies have been carried out to demonstrate the short-term or long-term effectiveness of NIMV in patients with the OHS. Today, it is accepted that NIMV is the treatment of choice of patients with kyphoscoliosis and respiratory failure.2 The aim of this study was to assess the effectiveness of NIMV in a group of patients with OHS and to compare the findings to those obtained in a group of patients with kyphoscoliosis also treated with NIMV. Materials and Methods Study Design This clinical trial compared patients at baseline and after 4 months of treatment with NIMV. There were two groups of Clinical Investigations

patients: one group with OHS and one group with kyphoscoliosis. The inclusion criteria for the patients with OHS were as follows: body mass index ⬎ 33, failure of dietetic treatment to induce weight loss with persistence of hypercapnic respiratory failure (Pco2 ⬎ 47 mm Hg) for at least 3 months before NIMV treatment, and refusal of surgical treatment of obesity based either on clinical criteria or by the patients themselves. The inclusion criteria for the kyphoscoliotic patients were as follows: scoliosis angle (Cobb) ⬎ 90° and hypercapnic respiratory failure (Pco2 ⬎ 47 mm Hg). Patients were excluded if the FEV1/FVC was lower than 0.65 or if the apnea/hypopnea index (AHI) obtained from a complete polysomnographic study was ⬎ 20 events per hour. Patients Thirty-eight patients were included initially: 23 patients in the OHS group and 15 in the kyphoscoliotic group. Two patients (one in the OHS group and another in the kyphoscoliotic group) rejected the treatment with NIMV. The remaining 36 patients completed the study protocol. All patients received oxygen therapy for at least 3 months, and they had been clinically stable for at least 3 months before the beginning of treatment with NIMV. Study Protocol A baseline clinical and pulmonary functional evaluation followed by an overnight polysomnographic study was obtained before NIMV treatment. If the sleep study revealed an AHI ⬍ 20, the patients were then admitted to the hospital to become familiarized with NIMV and to adjust the device to provide adequate ventilation with comfort. Once these goals were achieved, the patients were discharged home and followed up for 4 months. A new clinical and pulmonary function evaluation was performed after these 4 months. The study protocol was revised and approved by the local ethics committee. Measurements Clinical Parameters: The clinical outcomes (expressed as percentage of the total number of patients reporting the symptom) before and after NIMV are shown in Fig 1. They included the presence of morning headaches, morning drowsiness, sleepiness, dyspnea, and leg edema. Dyspnea was evaluated using the

modified Medical Research Council dyspnea scale, as cited by Mahler et al.5 We counted the number of patients who scored the two most severe grades of the scale at baseline and after NIMV. Sleepiness was defined according to the international classification of sleep disorders.6 A change in any lower degree of the scale was considered as improvement in sleepiness. All other outcomes were recorded as either present or absent. Respiratory Function: Pulmonary function test measurements were obtained with the patients in the sitting position and breathing room air. The forced spirometry readings were obtained using a pneumotachographic spirometer. Lung volume and airway resistance were determined using body plethysmography (2800 Transmural Body Box; Sensor Medics; Yorba Linda, CA). Arterial blood gases obtained without nasal ventilation in place were processed immediately in an analyzer (model 1306; Instrumentation Laboratory; Milan, Italy). Maximal inspiratory and expiratory muscle pressures were measured from residual volume and total lung capacity, respectively, using a manometer (163 Sibelmed; Sibel; Barcelona, Spain), after the method of Black and Hyatt.7 A small air leak prevented glottic closure, and the best of six efforts was recorded. In patients with kyphoscoliosis, height was estimated using the arm span.8 Overnight Polysomnography: Sleep studies using a commercially available system (Somnostard 4100; Sensor Medics) were obtained only at baseline to screen out patients with significant sleep apnea syndrome (AHI ⬎ 20). Sleep duration in every case was ⬎ 4 h. During sleep, the following parameters were recorded: electro-oculogram, ECG, electromyogram (submental and anterior tibial), EEG, oronasal flow with a thermistor, thoracic and abdominal movements with inductance plethysmographic bands (Respitrace Plus; NonInvasive Monitoring Systems; Miami Beach, FL), and oxygen saturation with a pulse oximeter. NIMV Treatment The in-hospital adaptation period for NIMV adjustment lasted from 3 to 7 days. Patients were treated initially with a volumecycled ventilator (Monal DCC; Taema; Paris, France), but in the patients who manifested poor tolerance to this system, we subsequently changed over to a bilevel pressure device (Onix Plus; Mallinckrodt SEFAM; Nancy, France). Five patients in the OHS group (23%) and one patient in the kyphoscoliotic group (7%) were treated with the bilevel device. An interface commercially available nasal mask (Modular Mask Sullivan; Resmed

Figure 1. Changes of clinical symptoms in patients with OHS and kyphoscoliosis when treated with NIMV (* ⫽ p ⬍ 0.05). CHEST / 119 / 4 / APRIL, 2001

1103

Limited; North Ryde, Australia) was used in all the patients. Initially, the ventilatory parameters were programmed to achieve the maximal reduction in Pco2, taking into account the patient’s tolerance and the air leakage. Finally, ventilatory adjustments were made to maintain the nocturnal oxygen saturation ⬎ 90%. If this saturation was not obtained, nocturnal oxygen was added. Statistical Analysis The significance of the difference between these determinations (basal and NIMV) was determined using ␹2 analysis or Fisher Exact Test for the percentage values. The numerical variables were analyzed using Student’s t test. If data did not have a normal distribution, a nonparametric test was used. A probability value of ⬍ 0.05 was deemed significant.

Results Follow-up Sixteen patients (73%) from the OHS group and 12 patients (86%) from the kyphoscoliotic group had been admitted to the hospital for respiratory failure at least once in the previous 3 years. During this time, the number (mean ⫾ SD) of emergency visits per patient was 5.2 ⫾ 2.1 visits for those in the OHS group and 4.6 ⫾ 1.7 visits for those in the kyphoscoliotic group. Neither emergency visits nor hospitalizations occurred during the follow-up period. All the patients were receiving NIMV treatment until the end of the follow-up period. Eleven patients in the OHS group (50%) and 8 patients in the kyphoscoliotic group (57%) required oxygen supplementation at the beginning of the study, to maintain nighttime oxygen saturation ⬎ 99%. The nighttime oxygen supplementation was progressively withdrawn in all patients, except for two patients in the OHS group and one patient in the kyphoscoliotic group. The compliance (mean ⫾ SD) measured by the respirator time counter at the end of the follow-up period was 7.2 ⫾ 0.8 h/d in the OHS group and 7.3 ⫾ 0.7 h/d in the kyphoscoliotic group. No patients increase their weight ⬎ 10% in the 4 months of follow-up. Clinical The anthropometric characteristics of these 36 patients are shown in Table 1. Patients in the OHS group were slightly older and had higher frequency of women than in the kyphoscoliotic group. Comparison of baseline symptoms between OHS and kyphoscoliosis patients revealed significant differences in two areas (Fig 1). More patients in the OHS group had sleepiness, while more patients in the kyphoscoliotic group reported severe dyspnea. However, after NIMV treatment, these differences were no longer significant. Improvements were reported in both groups after treatment with NIMV. 1104

Table 1—Anthropometric Characteristics of Patients Included in the Study* Characteristics

OHS

Kyphoscoliosis

Patients, No. Age, yr Male/female gender, No. BMI, initial BMI, end Cobb angle, degrees

22 61 ⫾ 14 6/18 40.8 ⫾ 8.0 40.9 ⫾ 7.0

14 43 ⫾ 20 7/7 24.8 ⫾ 8.0 25.1 ⫾ 7.0 111 ⫾ 15

*Data are presented as mean ⫾ SD unless otherwise indicated. BMI ⫽ body mass index

The change was statistically significant for all variables in patients with OHS and for all variables except sleepiness for patients with kyphoscoliosis (p ⬍ 0.1). Physiologic Figure 2 and Table 2 show the changes in physiologic outcomes. After NIMV treatment, there were statistically significant improvements in pH in the OHS group (p ⬍ 0.01) and in Po2 and Pco2 in both groups (p ⬍ 0.001). We did not find differences in the remaining physiologic parameters. When we compared pH, Po2, and Pco2 at the beginning and at the end of NIMV treatment between OHS and kyphoscoliotic patients, we did not find statistically significant differences. Discussion This study shows that NIMV is effective in alleviating the symptoms and reversing the respiratory failure of patients with OHS. The effect is similar to that seen in patients with kyphoscoliosis. The mechanism by which OHS causes respiratory failure has not been well established, although a central component characterized by a defect in the respiratory controller appears to be the most important in causing it.9 A decreased ventilatory response to CO2 rebreathing has been generally observed.10 In addition, some patients do not present an abnormal response to CO2, but decreased ventilatory response to hypoxemia.11 On the other hand, the fat deposit in the chest wall can modify the respiratory mechanism and increase the work of breathing.12 Decreases in lung volumes, chest wall and lung parenchymal compliances, and inspiratory muscle strength have been observed.13,14 NIMV treatment has been effective in improving respiratory failure with central origin3 or chest wall origin.2 In OHS, like in these diseases, NIMV used during sleep could revert the daytime respiratory failure probably by Clinical Investigations

Figure 2. Po2 and Pco2 evolution in patients with OHS and kyphoscoliosis when treated with NIMV. * ⫽ p ⬍ 0.001.

improving chemoreceptor sensitivity to Po2 and Paco2, although the increase of volume, lung compliance, and better respiratory muscle efficacy can be coadjuvant mechanisms.15 Today, it is accepted that NIMV is the treatment of choice of patients with kyphoscoliosis and respiratory failure,2 but there is little data related to OHS. In one study, during the application of NIMV, the gas exchange improved. In addition, there was a 40% decrease in inspiratory muscle load.16 Waldhorn17 treated three patients with OHS and apneas during sleep with conventional continuous positive airway pressure (CPAP). The respiratory failure did not improve in those three patients. In contrast, after 3 months of treatment with NIMV, there was a rever-

Table 2—Changes in Physiologic Outcome in OHS and Kyphoscoliotic Groups OHS Group Variables

Basal

NIMV

Kyphoscoliotic Group Basal

51 ⫾ 10 64 ⫾ 11† 53 ⫾ 6 Po2 Pco2 58 ⫾ 10 45 ⫾ 5† 59 ⫾ 11 pH 7.36 ⫾ 0.04 7.41 ⫾ 0.03‡ 7.35 ⫾ 0.03 FVC 61 ⫾ 12 62 ⫾ 13 28 ⫾ 9 FEV1 60 ⫾ 12 61 ⫾ 10 27 ⫾ 11 FEV1/FVC 80 ⫾ 13 81 ⫾ 10 84 ⫾ 8 TLC 86 ⫾ 13 87 ⫾ 15 48 ⫾ 8 RV 126 ⫾ 12 124 ⫾ 13 78 ⫾ 10 FRC 100 ⫾ 9 98 ⫾ 10 50 ⫾ 13 MIP 55 ⫾ 14 54 ⫾ 12 42 ⫾ 15 MEP 56 ⫾ 13 55 ⫾ 12 41 ⫾ 13

NIMV 65 ⫾ 5† 45 ⫾ 4† 7.39 ⫾ 0.02 30 ⫾ 12 30 ⫾ 12 85 ⫾ 11 49 ⫾ 9 77 ⫾ 11 51 ⫾ 11 42 ⫾ 14 41 ⫾ 11

*Data are presented as mean ⫾ SD. TLC ⫽ total lung capacity; RV ⫽ residual volume; FRC ⫽ functional residual capacity; MIP ⫽ maximal inspiratory pressure; MEP ⫽ maximal expiratory pressure. †p ⬍ 0.001. ‡p ⬍ 0.01.

sal of respiratory failure in the same three patients, as well as in a group of four kyphoscoliotic patients included in his series. Two subsequent studies18,19 treated 13 patients with pickwickian syndrome with CPAP. Similar to the results of Waldhorn,17 the treatment did not improve the respiratory failure in spite of being generally effective in abolishing the apneic events. The patients were then begun on NIMV treatment. With this treatment, the respiratory failure practically disappeared. Thereafter, the patients could be maintained out of respiratory failure by the use of simple CPAP. Similar results were obtained by treating six patients with PraderWilli syndrome with NIMV (hypercapnic respiratory failure, obesity, hypogonadism, learning difficulties, and frequently apneas during sleep).20 From these previous studies, it can be concluded that although NIMV treatment can prevent apneic events, the improvement of respiratory failure must be independent of this, as CPAP also prevented the apneic events but failed to revert the respiratory failure. One study4 has evaluated the effectiveness of NIMV on nighttime oxygen desaturation in 11 obese patients without daytime respiratory failure. This was compared with the effect of the “gold standard” treatment, nighttime oxygen. The results obtained were compared with a concurrent group of 10 patients with restrictive chest wall disease (mainly kyphoscoliotic) who received the same treatment protocol. In both groups, daytime Po2 improved significantly. Nighttime oxygen saturation and Paco2 also improved when the patients were treated with NIMV. Treatment with nocturnal oxygen only succeeded in improving nocturnal oxygenation, but in contrast to NIMV there was a tendency to increased nocturnal Paco2. One of the limitations in our present study was the lack of control group for patients with OHS. In all CHEST / 119 / 4 / APRIL, 2001

1105

the patients included in the OHS group, the standard treatment (oxygen therapy included) failed. As there was not a clear alternative treatment, then we thought it was not ethical to include a control group in this study. Another limitation of this study was that the follow-up period was not very long (4 months). Once the prospective follow-up of this study (4 months) was concluded, the patients were followed in a routine way. During the first year of treatment, one patient with kyphoscoliosis died (10th month). Another patient with obesity discontinued NIMV treatment because of facial herpes zoster and secondary neuralgia (ninth month). After 2 months of the withdrawal, the arterial blood gas levels were similar to those at the beginning of the treatment with NIMV. After 1 year of treatment, the remaining patients continued with NIMV treatment. During this first year, they did not attend the hospital or the Emergency Unit. The OHS patients suffered more frequently from daytime sleepiness than the kyphoscoliotic patients. Although some of these patients had some degree of obstructive sleep apnea, with an AHI ⬍ 20, it seems improbable that the resulting fragmentation of sleep could account for the daytime sleepiness, because of the fact that the number of arousals per hour of sleep was not different between obese patients and kyphoscoliotic patients. By this same reasoning, it is improbable that the obese patients presented upperairway increased resistance syndrome, although our polysomnograms did not include esophageal-pressure measurements. It is also possible that hypercapnia per se could induce sleepiness, but the observed levels were not different from those seen in the patients with kyphoscoliosis who reported less sleepiness. Therefore, it seems as if these obese patients develop daytime sleepiness because of causes different from intrinsic sleep alterations.21 Why they improve with NIMV treatment remains speculative. The first line of treatment of OHS is weight loss. This can be achieved with dietetic programs. As a rule, weight loss reverses the respiratory failure, and in addition treats the other diseases associated with obesity. However, the majority of patients with OHS do not lose enough weight, or if they do, they regain it rather quickly. Surgery for obesity is another therapeutic choice. Unfortunately, respiratory failure is considered an important factor that confers a bad prognosis in the upper-abdominal surgery of obese patients.22 Surgeons or anesthetists frequently reject the procedure, when obese patients manifest concomitant respiratory failure. One study has shown improvement in postsurgical pulmonary function when patients with gastroplasty are treated with NIMV during the first 24 h after surgery.23 This, 1106

coupled with our results, suggests that the surgical prognosis can improve if patients with OHS are treated preoperatively with NIMV. Perhaps the patients could then be considered for surgery once the respiratory failure has reversed. In summary, NIMV treatment improves the clinical symptoms and the respiratory failure of patients with OHS to a similar degree to that reported for diseases in which its use is completely established, such as kyphoscoliosis. The results of this study also suggest that the surgical prognosis in obese patients with hypercapnic respiratory failure could improve if those patients are treated preoperatively with NIMV. ACKNOWLEDGMENT: We are indebted to Vero´nica Rodrı´guez for assistance in the preparation of the article.

References 1 Metabolic and endocrine disorders. In: Gibson GB, ed. Clinical tests of respiratory function. London, UK: McMillan Press, 1984; 291–296 2 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 1994; 105:100 –105 3 Guilleminault C, Stoohs R, Schneider H, et al. Central alveolar hypoventilation and sleep: treatment by intermittent positive-pressure ventilation through nasal mask in an adult. Chest 1989; 96:1210 –1212 4 Masa JF, Celli BR, Riesco JA, et al. Noninvasive positive pressure ventilation and not oxygen may prevent overt ventilatory failure in patients with chest wall diseases. Chest 1997; 112:207–213 5 Mahler DA, Weinberg DH, Wells CK, et al. The measurement of dyspnea: contents, interobserver, agreement, and physiologic correlates of two new clinical indexes. Chest 1984; 85:751–758 6 International classification of sleep disorders: diagnostic and coding manual. Rochester, MN: American Sleep Disorders Association, 1990; 52–58 7 Black LF, Hyatt RE. Maximal respiratory pressures: normal value and relationship to age and sex. Am Rev Respir Dis 1969; 99:696 –702 8 Linnderholm H, Lidgren U. Prediction of spirometric values in patients with scoliosis. Acta Orthop Scand 1978; 49:469 – 474 9 Krachman S, Criner GJ. Hypoventilation syndromes. Clin Chest Med 1998; 19:139 –155 10 Sampson MG, Grassino A. Neurochemical properties in obese patients during carbon dioxide rebreathing. Am J Med 1983; 75:81–90 11 Zwillich CW, Sutton FD, Pierson DJ, et al. Decreased hypoxic ventilatory drive in the obesity-hypoventilation syndrome. Am J Med 1975; 59:343–348 12 Sharp JT, Henry JP, Sweany SK, et al. The total work of breathing in normal and obese men. J Clin Invest 1964; 43:728 –739 13 Thomas DS, Cowen ERT, Hulands G, et al. Respiratory function in the morbidly obese before and after weight loss. Thorax 1989; 4:382–386 14 Due DY, Bray G, Hansen JE, et al. Effects of obesity on respiratory function. Am Rev Respir Dis 1983; 128:501–506 15 Masa JF, Sa´nchez de Cos J, Disdier C, et al. Nasal intermittent positive pressure ventilation: analysis of its withdrawal. Chest 1995; 107:382–388 Clinical Investigations

16 Pankow W, Hijjeh N, Schuttler F, et al. Influence of noninvasive positive pressure ventilation on inspiratory muscle activity in obese subjects. Eur Respir J 1997; 10:2847–2852 17 Waldhorn RE. Nocturnal nasal intermittent positive pressure ventilation with bi-level positive airway pressure (BiPAP) in respiratory failure. Chest 1992; 101:516 –521 18 Piper AJ, Sullivan CE. Effects of Short-term NIPPV in the treatment of patients with severe obstructive sleep apnea and hypercapnia. Chest 1994; 105:434 – 440 19 Schafer H, Ewig S, Hasper E, et al. Failure of CPAP therapy in obstructive sleep apnoea syndrome: predictive factors and treatment with bilevel-positive airway pressure. Respir Med 1998; 92:208 –215

20 Smith IE, King MA, Siklos PW, et al. Treatment of ventilatory failure in the Prader-Willi syndrome. Eur Respir J 1998; 11:1150 –1152 21 Vgontzas AN, Bixler EO, Tan TL, et al. Obesity without sleep apnea is associated with daytime sleepiness. Arch Intern Med 1998; 158:1333–1337 22 Celli BR. Perioperative respiratory care of the patient undergoing upper abdominal surgery. Clin Chest Med 1993; 14:253–261 23 Joris JL, Sottiaux TM, Chiche JD, et al. Effect of bi-level positive airway pressure (BiPAP) nasal ventilation on the postoperative pulmonary restrictive syndrome in obese patients undergoing gastroplasty. Chest 1997; 111:665– 670

CHEST / 119 / 4 / APRIL, 2001

1107