Abnormalities of the Respiratory Function and Control of Ventilation in Patients with Anorexia nervosaGonzález-Moro J.M.R.a · de Miguel-Díez J.a · Paz-González L.b · Buendía-García M.J.a · Santacruz-Siminiani A.c · de Lucas-Ramos P.a
Departments of Pneumology,aHospital General Universitario Gregorio Marañón, Madrid, bHospital Morales Meseguer, Murcia, and cHospital Santa María del Rosell, Murcia, Spain Corresponding Author
Background: Anorexia nervosa is a good model to study the effects of malnutrition on the respiratory system, since it excludes the consequences that some chronic diseases can have on the lung. Objective: The objective of this study was to assess pulmonary function and alterations in the respiratory system in patients with anorexia nervosa. Methods: Twenty-two women, 12 with anorexia nervosa and 10 healthy volunteers, were studied. Anthropometric data were gathered for all the participants. In every case, an arterial blood gas test and functional respiratory study, that included spirometry, plethysmography, measure of maximum muscular respiratory pressure and study of the ventilatory pattern at baseline and after hypercapnic stimulation, were performed. Results: No significant differences were found in mean age and height in both groups of patients but there was a difference in body mass index (p < 0.05). In pulmonary function tests, an increase in residual volume (RV), RV/total lung capacity (TLC) ratio and functional residual capacity and a decrease in maximum respiratory pressure were found in patients with anorexia nervosa compared to the control group (p < 0.05). Differences were also found in the ventilatory pattern, with a reduction in minute ventilation and occlusion pressure as well as a decreased response of these parameters to hypercapnic stimulation (p < 0.05) in the patients with anorexia nervosa. Conclusion: In patients with anorexia nervosa, a significant elevation in RV, in the RV/TLC ratio and in functional residual capacity and a decrease in the maximum respiratory pressure were noted. In addition, they present an alteration in the central respiratory drive and a response of the respiratory system to hypercapnic stimulation. Although these alterations have no repercussion on the maintenance of gas exchange in baseline conditions, they may have in extreme situations.
Copyright © 2003 S. Karger AG, Basel
Morbid obesity and severe malnutrition are the two extremes in the spectrum of eating disorders. Repercussion of obesity on the different organs and systems has been widely studied in the past. However, the effect of serious malnutrition on the organism, especially on the respiratory system, is less known. Anorexia nervosa is a frequent cause of malnutrition and its prevalence has increased in recent years. It is a multicausal disorder, resulting from complex interactions between psychosocial and biological factors. Above all, it affects young women, generally without other diseases . Thus, anorexia nervosa is a good model to study the effects of malnutrition on the respiratory system, since it excludes the consequences that some chronic diseases can have on the lung. On the other hand, the existence of alterations in some neurotransmitters that can be involved in its pathogenesis or be a consequence of the malnutrition in the patients with this disorder has been described . Less attention has been given to the study of alterations in the ventilatory control in these patients .
This study aims to assess the effect of anorexia nervosa on respiratory function parameters and to analyze changes in the respiratory system in patients with this disease.
Patients and Methods
Patients with anorexia nervosa admitted to the Psychiatry Service of our hospital were prospectively evaluated in this study. All of them complied with the DSM-IV and ICD-10 criteria for the diagnosis of this disease. Subjects having a background of smoking or cardiorespiratory diseases and those who did not adequately collaborate in the performance of the respiratory function studies were excluded. A group of healthy volunteers recruited from the hospital personnel was assessed as a control. Both the patients as well as the control group were informed on the details of the examinations to be performed and gave their consent to their performance.
Clinical and anthropometric data of the subjects who participated in the study were gathered. Baseline arterial blood gas tests, simple spirometry, lung volume determination, measurement of maximum respiratory pressure and analysis of the ventilatory pattern at baseline and after hypercapnic stimulation were performed in all the patients. The arterial blood gas test was performed with the subject breathing room air and at rest, by puncture of the radial artery, with previous local anesthesia and immediate analysis in a gas analyzer (Electrode analyzer IL-1306, Instrumental Laboratory, Lexington, Mass., USA), according to the recommendations of the Spanish Society of Pneumology and Thoracic Surgery . Spirometric and plethysmographic maneuvers were performed with the patient in the sitting position, and the nose was occluded by a nose clip, according to the recommendations of the European Respiratory Society . One MasterLab Pro unit (Jaeger, Würzburg, Germany) was used for its measurement. Maximum inspiratory (Pimax) and expiratory (Pemax) pressures were measured with a manometer (163 Sibelmed, Sibel, Barcelona, Spain) following the classical method of Black and Hyatt . Pimax was measured from the residual volume, while the subject performed maximum inspiratory effort against a partially occluded obturator. Pemax was determined at total lung capacity (TLC) while the subject strongly exhaled against a partially occluded obturator. The highest value of three measurements in each of the maneuvers was taken as Pimax and Pemax. Occlusion pressure (P0.1) and the ventilatory pattern were also studied with MasterLab Pro equipment. The ventilatory pattern was measured at both baseline as well as during hypercapnic stimulation, produced on breathing into a closed circuit through a Douglas bag with a 7% mixture of CO2 and the rest of O2, following the Read  reinhalation method. To perform it, a circuit divided by a double pathway valve into an inspiratory part, which was connected to a Fleish-type pneumotachograph, and an expiratory one, in which the expired fraction of CO2 was continuously determined by means of a gas analyzer (Ergo-Oxyscreen, Jaeger), was used. P0.1 was determined according to the classic method of Whitelaw et al. . The subject breathed with a nasal clip and relaxed through a scuba-diver-type mouthpiece connected to the pneumotachograph, which was provided with the previously described bidirectional valve. In the inspiratory branch, an occlusion mechanism was placed that was intermittently and randomly activated at the end of an expiration, opening once the first 100 ms of inspiration have passed. Following the recommendations for the performance of this measurement, the laboratory setting was calm and without noise, the patient did not know the examination that was going to be done nor was he/she able to see the graphic recording, and the occlusion control was silent, leaving several breaths free between each occlusion. The final value of P0.1 was calculated, finding the mean of five reproducible maneuvers of correct morphology.
Statistical analysis was performed with the SPSS 9.0 statistical program for Windows. Descriptive data were expressed as means ± SD. Quantitative variables were compared by means of an unpaired t test. The Pearson correlation coefficient was applied when it was necessary to analyze the correlations between the quantitative data. Values of p < 0.05 were considered significant.
Twelve women with anorexia nervosa and 10 healthy women were enrolled. Table 1 shows the anthropometric characteristics of the study population. Weight and body mass index were significantly lower in the anorexia nervosa patient group (p < 0.05).
Table 1. Anthropometric data
Baseline blood gas parameters were normal in all the participants. No significant differences were found between the subjects with anorexia nervosa and the control group in arterial O pressure (PaO2; 87.0 ± 7.4 vs. 94.4 ± 4.2 mm Hg, respectively) and arterial CO2 pressure (PaCO2; 38.6 ± 4.4 vs. 35.1 ± 7.2 mm Hg, respectively; fig. 1). However, the individual analysis showed that no healthy individual had baseline values of PaCO2 superior to 40 mm Hg, while PaCO2 was >40 mm Hg in 5 women with anorexia nervosa.
Fig. 1. Blood gas test values before and after hypercapnic stimulation. No significant differences between the baseline values of the patients with anorexia nervosa and the control group or between the post-stimulation values in both groups were found. AN = Anorexia nervosa; C = control.
Regarding the respiratory function tests, spirometry was normal in all the subjects evaluated. Patients with anorexia nervosa presented significantly higher values of residual volume (RV; 146.6 ± 47.6 vs. 94.0 ± 21.5%, p < 0.05), RV/TLC ratio (35.5 ± 11.3 vs. 25.5 ± 6.8, p < 0.05) and functional residual capacity (129.5 ± 23.1 vs. 93.4 ± 10.9%, p < 0.05) than the control group. Furthermore, patients with anorexia nervosa showed a significantly lower Pimax than the healthy volunteers (63.3 ± 13.4 vs. 91.0 ± 14.9 cm H2O, respectively, p < 0.05). Significant differences were also detected in the Pemax in both groups (82.0 ± 19.8 vs. 116.3 ± 22.5 cm H2O, respectively, p < 0.05). In the control group, the individual data showed only 2 patients with Pimax inferior to 80 cm H2O and none with Pemax below that value. In the group of women with anorexia nervosa, 10 patients were found with Pimax lower than 80 cm H2O and 5 with Pemax being lower than that value. Following correction of lung volumes, Pimax and Pemax remained reduced. On the other hand, no differences were found in the remaining parameters (table 2).
Table 2. Respiratory function parameters
Table 3 shows the ventilatory parameters. Significant differences were detected in tidal volume, minute ventilation (VE) and occlusion pressure (P0.1) in both groups. A significant correlation was also found between body mass index and minute volume in the patients with anorexia nervosa (r = 0.705; p < 0.05). On the other hand, the ventilatory response to hypercapnic stimulation was significantly lower in these patients in relationship with the controls (table 4). The individual values of PaCO2 and ventilatory response to hypercapnic stimulation can be seen in table 5. Excluding patients with PaCO2 >40 mm Hg, the ventilatory response remained reduced. A significant correlation was also found between the PaCO2 value after hypercapnic stimulation and both the ΔVE/ΔPaCO2 slope (r = –0.692; p < 0.05; fig. 2) and the ΔP0.1/ΔPaCO2 slope (r = –0.558; p = 0.05) in the subjects with anorexia nervosa, consequently, the ventilatory response to hypercapnia decreased with increasing post-stimulation PaCO2.
Table 3. Ventilatory pattern
Table 4. Ventilatory response to the hypercapnic stimulation
Table 5. Individual values of PaCO2 and ventilatory response to hypercapnic stimulation
Fig. 2. Correlation between the VE/PaCO2 ratio and PaCO2 following hypercapnia in the patients with anorexia nervosa.
In the first place, our study demonstrates that patients with anorexia nervosa present air trapping and lung hyperinflation, possibly in relationship with respiratory muscular dysfunction caused by extreme malnutrition. Second, these patients manifest a decrease in VE, since their ventilatory pattern is slower and more superficial, and a decrease in central respiratory impulse and ventilatory response to chemical stimulation.
Malnutrition can cause important adverse effects on the respiratory system. A deficient nutrition status can worsen respiratory muscular function , alter ventilatory impulse drive , contribute to the development of pulmonary parenchymatous lesions  and deteriorate respiratory defense mechanisms . These effects have been studied mainly in patients with chronic diseases with systemic involvement, so that the role of factors other than malnutrition on respiratory muscular dysfunction cannot be excluded.
In this study, young undernourished patients with anorexia nervosa with no other coexisting disease have been evaluated. Thus, it was possible to assess the real role of malnutrition on the respiratory function. In some previous studies performed in patients with this disease, a decrease in FVC, increases in RV and the RV/TLC ratio and a decrease in maximum respiratory pressure have been observed, all in relation with respiratory muscular weakness caused by extreme malnutrition [13, 14, 15]. In this sense, a marked decrease in muscular force of the diaphragm has been described in these patients, with improvement after nutritional replacement . Among the mechanisms responsible for muscular dysfunction presented by patients with malnutrition, alterations in the size, distribution and myofibrillar content of muscular fibers, reduction in the incorporation of amino acids to the muscles and alterations in the protein composition and oxidative enzyme activity are found . On the other hand, it has also been suggested that patients with anorexia nervosa present a loss of elastic properties of the lung tissue, since isolated cases of subcutaneous emphysema and pneumomediastinum have been recorded after cough maneuvers, vomiting or extreme exercise [17, 18]. However, in those studies in which this aspect has been evaluated, no decrease has been found in the diffusion capacity for carbon monoxide, which constitutes the parameter that best correlates with the histologic diagnosis of emphysema . The results obtained in this study, in which a control group is also included, verify that there is an increase in RV, RV/TLC ratio and functional residual capacity and worsening of respiratory muscular function in patients with anorexia nervosa compared to the control group. On the contrary, no significant spirometric or blood gas test alterations were detected in these patients, and the diffusion capacity for carbon monoxide was also not evaluated. It is possible that the reduction in oxygen consumption, CO2 production and energetic output present in subjects with anorexia nervosa can delay the onset of hypercapnia [3, 19]. In any event, this blood gas test alteration does not generally appear before the respiratory muscular force reaches a value inferior to 30% of its predicted value. On the other hand, it has been stated that the alteration described can improve with nutritional replacement .
In our study, it was also seen that patients with anorexia nervosa present a slower and more superficial respiratory pattern. Intensity of this alteration is greater as the weight loss increases. In addition, these patients manifest a reduction in the central respiratory impulse, as is shown by the fact that they present a reduction in P0.1, a parameter that measures the negative pressure produced in the mouth in the first 100 ms, as a consequence of the contraction of the inspiratory muscles, when an inspiration against the occluded airway is performed. Although this alteration could be justified, at least partially, by the decrease in respiratory muscle strength, such an important reduction as that found in P0.1 suggests that there is an added component of alteration in the neurons that govern the respiratory center. On the other hand, the ventilatory response to hypercapnic stimulation, both of P0.1 as well as VE, is reduced in these patients compared to the control group. This alteration is important since it has been described that the response of ventilation to chemical stimulation tends to remain preserved, even in the presence of a marked weakness of the inspiratory muscles [20, 21]. Severe malnutrition would therefore be the mechanism responsible for the decreasing ventilatory response to hypercapnia detected in the patients with anorexia in our study.
One limitation of this study is that we did not have radiological studies or diffusion indexes to detect or exclude emphysematous changes in the lungs secondary to malnutrition. Most of the findings discovered (hyperinsufflation and decreased maximum respiratory pressure) may be caused by emphysema. However, the previous studies performed in humans do not support the hypothesis that anorexia nervosa leads to the development of emphysema, even in smokers .
The results of this study have marked clinical implications. As a consequence of all the alterations described, patients with anorexia nervosa who are malnourished may have a greater predisposition to suffering respiratory problems . The existence of alterations in pulmonary immunocompetence has been described in these patients and could favor the development of infections [23, 24]. The additional problem that respiratory-muscle weakness could make cough maneuvers ineffective is added to this.
In conclusion, the results obtained reflect the adverse effects of anorexia nervosa and malnutrition on the respiratory system. Among these, air trapping, lung hyperinflation, dysfunction of the respiratory muscles and reduction in the respiratory impulse and ventilatory response to hypercapnia stand out. Although these alterations do not have an important repercussion on gas exchange in these subjects under baseline conditions, they may exert a greater impact on the respiratory function in extreme situations.
Servicio de Neumología
Hospital General Universitario Gregorio Marañón
C/ Doctor Esquerdo, 46, ES–28007 Madrid (Spain)
Tel. +34 91 5868330, Fax +34 91 5868018, E-Mail email@example.com
Received: August 29, 2002
Accepted after revision: April 24, 2003
Number of Print Pages : 6
Number of Figures : 2, Number of Tables : 5, Number of References : 24
Respiration (International Review of Thoracic Diseases)
Founded 1944 as ‘Schweizerische Zeitschrift für Tuberkulose und Pneumonologie’ by E. Bachmann, M. Gilbert, F. Häberlin, W. Löffler, P. Steiner and E. Uehlinger, continued 1962–1967 as ‘Medicina Thoracalis’ as of 1968 as ‘Respiration’, H. Herzog (1962–1997)
Official Journal of the European Association for Bronchology and Interventional Pulmonology
Vol. 70, No. 5, Year 2003 (Cover Date: September-October 2003)
Journal Editor: C.T. Bolliger, Cape Town
ISSN: 0025–7931 (print), 1423–0356 (Online)
For additional information: http://www.karger.com/journals/res