INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: positive-pressure respiration; quality of life; respiratory insufficiency; sex behavior

Although age and gender have an impact on sexual behavior (1), sexualityas a central issue of quality of lifelasts a lifetime. In general, health status is an important predictor of male sexual activity but less influential for female sexual activity, where the existence of a sexually interested partner and a pleasurable sexual biography are even more important (12).

In patients with chronic lung disease or respiratory muscle weakness, the fear of dyspnea and reduced exercise tolerance often limit the sexual activity of patients (18). Furthermore, misconceptions, ignorance, and poor physical or mental health are all common in this population and contribute to sexual dysfunction (22).

Independent of the underlying diagnosis permanent hypercapnia represents a late stage of chronic lung diseases. It has been shown that life expectancy is markedly reduced in patients with chronic hypercapnia (25, 26). Furthermore, daily activity and quality of life are reduced in chronic hypercapnic respiratory failure (CRF) (27). One may speculate that sexual activity is also compromised in patients with CRF.

Noninvasive mechanical ventilation (NMV) can improve daytime arterial gas tension in CRF, especially in patients with extrapulmonary restrictive disorders and neuromuscular disease in whom NMV is generally well tolerated and seems to be associated with an improved prognosis (28).

Another goal of NMV is to enhance quality of life. It has been shown that NMV causes an improvement in self-reported quality of life in patients with CRF (27, 29, 30).

Systematic searches in medical/scientific databases (Medline, 1966-2001; Web of Science version 4.1, February 2000 to 2001) for publications referring to "chronic hypercapnic respiratory failure or noninvasive mechanical ventilation" and "sexual behavior or sexual problems" resulted in no references researching human patients. Because this topic has not been studied before, we have conducted the first study of the sexuality of patients using long-term NMV at home.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The questionnaire was sent to 383 patients with NMV due to CRF. The diagnoses and devices used by the whole population are given in Table 1.

Study Protocol

For each patient the adaptation to NMV had been completed in our hospital (Krankenhaus Kloster Grafschaft, Schmallenberg, Germany), which is a supraregional center for mechanical ventilation. After adaptation to NMV all patients were readmitted every 6 mo to assess their physiologic condition and, if indicated, to readjust the ventilator.

In the meantime all patients received care from their chest physicians and were admitted to our hospital if needed. We were informed by the patients' chest physicians about those who died after the last hospital stay.

Patient lung function data, blood gases at rest, and, if available, exercise data were taken from the records generated during the most recent visit to our hospital. The ergometer tests were performed at the pre-NMV maximal work rate in a constant load protocol. In patients with severe dyspnea already at rest or those who were not able to cycle because of other reasons (e.g., orthopedic or neuromuscular disability involving the lower extremities), ergometry was not performed.

The study protocol was approved by the institutional review board.

Questionnaire

The enrolled patients were sent the questionnaire on March 1, 2000 accompanied by a letter describing the study objectives. The questionnaire was constructed to gather factual information about demographic data, social data, technical details concerning the interface, sexual activity, and changes in sexual activity after initiation of the NMV. Furthermore, we also used open-ended questions to elicit as broad as possible a range of information concerning sexual behavior.

The questionnaire items were as follows:

1. Age

2. Gender

3. Marital status

4. Relationship status (with partner; without partner/single)

5. How long have you used your ventilator?

6. Which type of mask do you use?

• Nasal mask
• Face mask

7. Are you sexually active?

8. If you are sexually active, how did your sexual behavior change with the use of the ventilator compared with the time before ventilation?

• Increased sexual activity
• Increased imagination
• No change
• Decreased sexual activity

9. If you are sexually active, how often do you practice sexual intercourse per month?

10. Is sexual intercourse possible without the ventilator?

11. Do you practice masturbation?

12. If you are not sexually active (any more) or your sexual activity decreased, what are the reasons for this?

• I am too old
• The device is annoying
• I am too sick
• Other: _________________

13. If you feel disturbed by the device during sexual activities, what is most annoying?__________________

In light of the intimate topic of our study, data were collected anonymously.

To correlate physiologic parameters with questionnaire responses, the patients' lung function data, blood gases, and ergometry test data were attached to the questionnaire at all times in the process. The individual ventilation mode, diagnosis, or further details were not recorded.

Because some questions were not answered by all patients, the sample size varied during analysis.

Inclusion Criteria

All patients demonstrated slow progression of the underlying disease leading to CRF (i.e., patients with chronic obstructive pulmonary disease [COPD], thoracic restriction, and neuromuscular diseases).

All patients tolerated NMV (i.e., mask and setting of the ventilator) and NMV was effective (i.e., normocapnia during daytime mechanical ventilation). All patients had been treated with NMV for at least 6 mo and using NMV for more than 5 h/d. The last in-hospital visit was not more than 6 mo before study enrollment. The patients had to be clinically stable at the last hospital visit and after discharge (observed by their own chest physician). Only patients older than 40 yr were included.

Exclusion Criteria

Patients were excluded for rapidly progressive neuromuscular diseases (e.g., progressive amyotrophic lateral sclerosis), invasive mechanical ventilation via tracheotomy, acute respiratory failure or acute exacerbation at the last control visit and after discharge (observed by the chest physician), lack of cooperation, mental disability, or participation in an exercise rehabilitation program.

Measurements

Body plethysmography was performed with a Masterlab (E. Jäger, Würzburg, Germany). Normal values for standard lung function data were taken from Quanjer and coworkers (31).

The samples for capillary blood gas analysis (Gas Check; AVL, Bad Homburg, Germany) at rest were obtained from the earlobe without oxygen supplementation.

The cycle ergometer test was performed on an electronically braked cycle ergometer (Ergoline; Siemens, Erlangen, Germany). The maximal work load of the ergometry test was determined before starting NMV, using an incremental protocol. The ergometry test was terminated either by the patient (symptoms of fatigue or breathlessness), or by the operator if the patient failed to maintain the given pedal frequency (50 revolutions/min). The performance time was recorded.

Ventilation and Interface

The intervention period began with a run-in period of NMV for several days to determine the optimal ventilator settings. Initially all patients underwent volume-targeted ventilation (VTV). If the patients felt uncomfortable, VTV was replaced by pressure-targeted ventilation (PTV). Further details on adaptation to both VTV and PTV have been described elsewhere (32).

All patients were initially ventilated via a conventional nose or full face mask (Respironics, Murrysville, PA; and Resmed, Sydney, Australia). If pressure sores developed during the adaptation or follow-up or if excessive leak around the mask was detected, a customized nose or full face mask was manufactured.

Normal Collective and Compared Issues

Data from normal subjects (> 45 yr old) were acquired by a survey dealing with sexual behavior of German-speaking inhabitants of Switzerland. This project was initiated by the interdisciplinary research group for sex studies (University of Zurich, Zurich, Switzerland). The data collection was performed in the first half of 1999 and was based on a standardized questionnaire. In total, 3,418 questionnaires were sent and 1,498 subjects (641 male, 857 female) answered (i.e., 43.8%). The mean age of male respondents was 61.8 yr and that of female respondents was 58.2 yr. Results were published in 2001 (33).

On the basis of both databases four main categories were compared between patients and normal subjects: (1) percentage sexually active, (2) percentage performing masturbation, (3) frequency of intercourse, and (4) reasons for sexual inactivity.

Statistics

Results are expressed as means ± standard deviation (SD). For interindividual tests between groups, t tests for independent samples were used if assumptions of normal distribution and homogeneity of variances were met; Mann-Whitney U tests were used otherwise. For analysis of frequency of coitus by diagnostic category, Kruskal-Wallis analysis of variance (ANOVA) was used. Frequency tables (2 × 2) were evaluated by Fisher exact tests. To detect different response rates between age classes or diagnostic categories, observed and expected frequencies were compared by ² test. Correlations between two variables were tested by Spearman rank correlation. Comparisons between patients and normal subjects were performed with tests for difference between two percentages (z test) in the case of categorical variables and by Mann-Whitney U test in the case of frequencies of coitus. In all cases, a two-tailed p value < 0.05 was considered significant. The data were analyzed with STATISTICA (34).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Anthropometric and Physiologic Data

Of 383 questionnaires sent to the patients by mail, 54.3% were returned. Thus 208 patients (142 male, 66 female) were included in the study. There were no differences between respondents and the whole population regarding blood gases or lung function data (p > 0.2 in all cases), age distribution (p = 0.43), or diagnostic categories (p = 0.52). The response rate was similar for males and females (55.9 versus 51.2%; p = 0.70).

Table 2 gives anthropometric data (age and gender) and social status (married, with a partner, or single) of responders and their Pa_O2, Pa_CO2, pH at rest (breathing room air), vital capacity (VC), and forced expiratory volume in 1 s (FEV₁). In total, 115 patients performed the ergometer test. The endurance time was 4.2 ± 2.0 min at a maximal work load of 65.2 ± 34.2 W (predicted, 142.2 ± 56.5 W).

NMV-associated Details

At the time of survey, subjects had received NMV for 41.1 ± 27.0 mo. A full face mask was used by 24% (n = 50 patients); a nasal mask was applied by 76% (n = 158 patients).

Sexual Activity

Of all patients, 34.1% were sexually active, 61.1% were not active, and 4.8% of the patients did not answer this question. We found no significant difference between males (37.9% sexually active) and females (27.3%; p = 0.20). An inverse relationship was found between sexual activity and age (p < 0.001), the most active group was the cohort between 40 and 50 yr old (68.2% sexual active; Figure 1). Between 50 and 60 yr of age the probability of being sexually active was higher for males than for females (54.3 versus 25.0%; p = 0.041). Sexually active patients had greater VC (2.1 versus 1.8 L; p = 0.042), higher absolute FEV₁ (1.4 versus 1.1 L; p = 0.002), and a higher Pa_O2 at rest (64.0 versus 60.4 mm Hg; p = 0.019). They also reached a higher maximal work load during exercise (72.0 versus 58.8 W; p = 0.044). We found no difference between main diagnostic categories (p = 0.26).

View larger version (74K): [in this window] [in a new window]   Figure 1.   Percentage of sexually active patients in different age classes. Sexual activity decreases with increasing age (p < 0.001). More males than females were active in the cohort of patients between 50 and 60 yr of age (p = 0.041).

Sexual activity was associated with social status. Compared with single patients, a higher percentage of patients who were married or with partners were sexually active. We found no difference between married patients and those with partners (p = 0.78), but in both groups the percentage of sexually active persons was greater compared with the single group (p < 0.002; Figure 2). There were no significant gender differences in the married and the related group but only male singles were sexually active (p = 0.046).

View larger version (76K): [in this window] [in a new window]   Figure 2.   Percentage of sexually active patients with different partner status. The sexual activity of single patients was reduced (p < 0.002) and found only in males (p = 0.046).

Compared with normal subjects the percentage of sexually active patients was markedly reduced (34.1 versus 84.0%; p < 0.001), independent of gender. This difference continuously increased with increasing age. The greatest differences were found in the cohort of patients older than 70 yr (15.1 versus 63.3%; p < 0.001) and in singles (6.8 versus 61.3%; p < 0.001).

Changes in Sexual Activity

Changes in sexual activity after NMV initiation were reported to be as follows: "no change," 46.3%; "decreased activity," 35.8%; "increased activity," 12.6%; and "increased fantasy," 10.5% (multiple answers possible).

No correlation was found between responses to "change of sexual activity after initiation of NMV" and blood gases, lung function, endurance, anthropometric data, or main diagnostic category. Those patients who indicated increased sexual activity had been undergoing NMV for 57.3 mo whereas those who described "no change" had been undergoing NMV for 43.2 mo. However, this trend did not reach significance (p = 0.093). Increased sexual fantasy correlated significant with sex; it was found only in males (p = 0.031).

Sexual Intercourse

Patients who classified themselves as "sexually active" had sexual intercourse 5.4 ± 4.8 times per month (range; 0-30 times per month). There was a negative correlation between frequency of sexual intercourse and age of patient (r_S = 0.403; p = 0.001). No other correlations were found, and again there was no difference between main diagnostic categories (p = 0.29).

Compared with normal subjects the frequency of coitus within the whole group of sexually active patients was increased (5.4 ± 4.8 versus 3.8 ± 4.3 times per month; p < 0.001), regardless of gender. We found the same difference in married patients (5.3 ± 4.6 versus 3.8 ± 3.7 times per month; p < 0.001) and an even greater difference in the cohort younger than 50 yr (7.9 ± 3.6 versus 4.9 ± 4.6 times per month; p < 0.003), but not in the group with unmarried partners (5.3 ± 5.1 versus 6.3 ± 6.0 times per month; p > 0.7).

In the majority of sexually active patients, intercourse was possible without the respirator. Four patients were ventilator dependent during intercourse, all were male, all were married, and all used a nasal mask during intercourse.

Masturbation

Masturbating patients were younger compared with nonmasturbating patients (58.0 versus 63.8 yr; p = 0.003).

In the sexually active group another factor influencing masturbation was relationship status: in single patients the percentage denying masturbation was significantly lower compared with the married group and the group with partners (p < 0.03; Figure 3). We found no significant gender difference (p > 0.48).

View larger version (68K): [in this window] [in a new window]   Figure 3.   Percentage of masturbating patients in the sexually active group. Masturbation was found mainly among single males (p < 0.03).

Compared with normal subjects the percentage of masturbating patients was markedly reduced (13.0 versus 39.8%; p < 0.001), regardless of gender.

The difference increases with increasing age from comparable values in the youngest cohort (< 50 yr) (31.8 versus 46.5%; p > 0.18) to extreme differences in the oldest cohort (> 70 yr) (5.7 versus 27.1%; p < 0.002). We found no significant differences in the group with unmarried partners (21.4 versus 43.5%; p > 0.10), but did in the married group (11.6 versus 36.8%; p < 0.001) and single patient group (15.9 versus 47.3%; p < 0.001).

Reasons for Sexual Inactivity

Reasons for sexual inactivity or decreased activity were as follows: "too sick," 31.3%; "too old," 18.8%; "no partner," 7.2%; "device is annoying," 4.8%; "breathlessness," 4.8%; "others," 17.3%. (Multiple answers were possible for this question and some patients did give multiple reasons.)

The answer "I am too sick" correlated with a lower PO₂ at rest (62.7 versus 58.9 mm Hg; p = 0.015) and reduced FEV₁ as a percentage of VC (FEV₁%VC) (59.4 versus 67.4%; p = 0.002). It was predominantly given by males (36.7 versus 21.9%; p = 0.034).

The reason "I am too old" for reduced sexual activity correlated with higher age (69.9 versus 61.6 yr; p < 0.001) and reduced ergometer endurance time (3.46 versus 4.32 min; p = 0.023).

The answer "missing partner" as the cause of reduced sexual activity correlated with status (single versus married or with partner; p < 0.001), gender (female versus male; p < 0.001; Figure 4), and FEV₁%VC (74.3 versus 64.1%; p = 0.023). "Missing partner" becomes prominent in singles, females, and patients with high FEV₁%VC.

View larger version (67K): [in this window] [in a new window]   Figure 4.   Percentage of patients answering "sexual activity reduced because of missing partner"; most often found in single females (p < 0.001).

Irritation caused by the respirator was not associated with any other variable.

In comparing sexually inactive patients and sexually inactive healthy subjects concerning their reasons for inactivity, we found similar percentages answering "too old" (28.4 versus 23.21%; p > 0.28). In patients the reason "too sick" was markedly increased (41.7 versus 8.0%; p < 0.001) whereas the reason "missing partner" was decreased (11.0 versus 52.7%; p < 0.001).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study is the first to study sexual activity and sexual problems in patients with NMV due to CRF.

One main finding of this study was that 34.1% of patients with NMV due to CRF were sexually active. Compared with the sexually inactive group, the active population was characterized by better lung function, younger age, and living with a spouse or partner (which is confounded with male gender).

Similar to other studies, we found that the sexual activity of patients with CRF is related to the same factors that are important for "normal" or healthy middle-aged or older persons. Namely, gender, partner status (single or with partner), and age (1, 35).

However, compared with normal subjects the sexual activity and masturbation rate of patients with NMV are markedly reduced in the total population and decrease further with growing age of the patients. This is in line with the literature. Thus, according to the judgment of patients, 80% of patients attribute reduction of sexual activity to somatic effects of a chronic lung disease (36). Sexual activity causes an increase in tidal volume and breathing frequency (36) that is associated with the increased cardiopulmonary load. The energy cost of human physical activities has been classified (19, 37). "Metabolic equivalents" (Met) have been defined as oxygen cost per kilogram per minute at rest. The pre- and postorgasm period needs 3-4 Mets (i.e., walking 3 mi). The maximal energy expenditure during an orgasm lasting 30 s is 5-6 Mets (i.e., walking 4 mi/h, cycling 10 mi/h, or continuously climbing stairs for 4-6 min). Thus, in patients with chronic lung diseases, sexual activity may often be reduced because of exertional dyspnea.

The second main finding was that with a mean value of 5.4 episodes of intercourse per month sexually active patients were even slightly more active than normal subjects of the same age. This fact was also found consistently in different age categories. One explanation could be that NMV selects a subgroup of patients who tend to be more active sexually and enables the patients to realize it. Even if we did not evaluate the psychological background these patients might be more optimistic (38) or have a higher-level internal locus of control (39).

This hypothesis is supported by the next finding, that NMV influenced sexual activity in more than 50% of our group of patients. However, this influence went in both directions: in approximately half the patients, either sexual activity or imagination increased, whereas in the other half sexual activity decreased after NMV was initiated.

As has been said before, sexuality induces energy expenditure and loads the cardiorespiratory system. Even if the questionnaire did not elucidate further details concerning this area, one may speculate that the subgroup with increased sexual activity profited from the improved physical activity that may occur after initiation of NMV (40). Some physiological factors discriminating between active and inactive group (VC and FEV₁) are not significantly influenced by NMV, according to the literature (43). But two factors, PO₂ at rest and exercise tolerance, correlated with reduced sexual activity may be improved by NMV (19). Furthermore, we found an improvement of hemodynamics caused by long-term application of NMV (44), which also may influence physical and sexual activity.

In the majority of patients, intercourse was possible without using the ventilator simultaneously. However, four male married patients were ventilator dependent, with a nasal mask, during intercourse. The authors had the opportunity to interview two patients (48 and 55 yr old) in this group. Both suffered from severe thoracorestrictive disorders. Before initiating NMV intercourse had not been possible for many years because of excessive dyspnea and physical weakness. Both patients stated that their partners were actively involved in handling the interface, head gear, and even the adjustment of the ventilator settings during intercourse. To reach a sexual climax, both patients needed higher frequency and tidal volume settings during intercourse compared with the usual setting of the ventilator setting at rest.

It remains a provocative finding that half the patients experienced a decrease in their sexual activity after the initiation of NMV. The majority of this subgroup judged themselves either "too sick" or "too old," which was confirmed by the correlation with the reduced PO₂ at rest and reduced FEV₁ ("too sick") and the higher age and reduced endurance time ("too old").

According to the long time interval between initiation of NMV and answering of the questionnaire (41.1 ± 27.0 mo) it is not possible to separate the NMV-associated negative effects from the spontaneous progression of the underlying disease. The included diseases are chronic and progressive and therefore the decrease in sexual activity after initiation of NMV could be due to progression of the disease rather than NMV itself.

Besides a control group without NMV, a "before and after NMV" comparison would have been possible to obtain more insight into the intraindividual development depending on NMV. However, we chose an anonymous one-point study design, in order to obtain the maximal responder rate and frank answers, because interviews we did in a pilot study dealing with sexuality in patients with CRF proved inadequate in obtaining maximal information about this delicate topic.

Despite the fact that most patients answered "I am too sick" or "I am too old" as the reason for absent or reduced sexual activity, the factor that discriminated best between the active and inactive group was partner status, that is, the presence of a spouse or partner. Other reasons for reduced sexual activity may be based on esthetic issues. One may speculate that the patient's partner may be irritated by the noise of the ventilator and the interface.

There are several limitations to this study. The ideal control group would have consisted of patients with CRF but without NMV. However, it seemed not to be ethically justifiable to us to withhold NMV from patients with hypercapnic respiratory failure.

We chose to compare our study population values with the "normal" values of an age- and gender-matched population. Because about two-thirds of our questionnaire dealt with NMV-specific topics or changes in sexuality caused by NMV it would therefore not be applicable to normal or disabled persons without NMV. Thus, general issues were compared concerning sexual activity, number of sexual intercourses per month, masturbation, and reasons for sexual inactivity. These data were collected in 1999 from a German-speaking population in Switzerland (33).

The response rate of approximately 50% may have caused selection bias. It is possible that patients who were sexually inactive and/or were single tended not to respond. However, a responder bias by age, gender, diagnostic category, or severity of disease could be excluded. The questionnaire was not administered during an interview. There were a number of missed questionnaire items and this might have been reduced by an interview format. Furthermore, one question (no. 8) may have caused confusion, as it was unclear to some patients which time interval was asked for (did "before ventilation" also mean before illness?); in addition, we did not include the option "decreased imagination" in Question 8. The advantages of an interviewer-based approach would have to be weighted against the possibility of a reduction of study recruitment.

More engagement of therapeutic teams dealing with chronically ill patients is needed to obtain information about patient sexuality. Applying general or disease-specific questionnaires dealing with quality of life would be an adequate means to investigate this area. However, only a few questionnaires deal with sexuality (45, 46). Therefore sexuality must be integrated and validated for quality of life questionnaires.

Another crucial issue concerns which branch of medicine should address the sexuality of patients with chronic and pulmonary diseases. Despite its general approach, rehabilitation medicine is predisposed to focus on patient sexuality. However, a survey dealing with activities and training programs of 44 Canadian rehabilitation units showed that only 47% dealt with this topic (47). Rehabilitation medicine in general, but also departments of pulmonary diseases, should realize their responsibility in this field.

In conclusion, we found that sexuality in patients receiving NMV for CRF is markedly reduced compared with that of normal subjects. In half of the patients, sexual activity was influenced by initiation of NMV.