Effect of Octreotide, a Somatostatin Analog, on Sleep Apnea in Patients with Acromegaly

Methods

Patients

The study group consisted of 19 patients (14 men and 5 women; mean age, 50 ±2 years [range, 26 to 68 years]) with active acromegaly. These 19 patients were part of a larger group of consecutive patients (n = 29) with clinically and biochemically active acromegaly who were assessed at the endocrine clinic at Saint Vincent's Hospital from 1987 to 1991 and who participated in a prospective trial assessing the efficacy of octreotide. Data from this larger group of patients have previously been reported in part [12]. Only 4 other patients with active acromegaly assessed during the period were considered ineligible or did not consent to octreotide treatment. Reasons for ineligibility included presence of gallstones and recent cholecystitis, a possible history of intravenous drug abuse, or failure to give informed consent.

All 29 patients had a complete sleep study, and 19 patients with sleep apnea (defined as 5 or more apneas/h) were entered into the study protocol for follow-up sleep studies. Ten patients were excluded either because they had no sleep apnea at baseline or because they were geographically isolated and it was very difficult for them to travel to the sleep study center. One of the 19 patients was included because, despite his having only 2 apneas per hour, we were interested in whether his degree of sleep apnea would be exacerbated by octreotide treatment. The diagnosis of acromegaly was made on clinical grounds (soft-tissue enlargement, facial disfigurement, and joint pain) and on biochemical grounds (increased IGF-1 levels and nonsuppressible growth hormone levels in response to an oral glucose load). The baseline sleep and hormonal data on 12 of the 19 patients have been reported previously [8, 12].

Twelve patients had not been previously treated. Six patients had previous pituitary surgery, 5 of whom received additional megavoltage irradiation. All 6 had persisting residual active disease that was clinically significant, with mean levels of growth hormone and IGF-1 above the reference range (see below). One patient had been unsuccessfully treated with bromocriptine (no change in levels of growth hormone and IGF-1). The drug was withdrawn 2 months before entry in this study. All surgically treated patients received stable thyroxine replacement, which was maintained throughout the study.

Study Protocol

All 19 patients had sleep studies at baseline and after 6 months of octreotide therapy (it was not logistically possible to study all of the patients in the first week). The first 8 patients also had sleep studies done during the first week of treatment to evaluate the short-term effect of octreotide. This early phase of evaluation was terminated when it became apparent that no clinically significant adverse effects occurred (see Results). A subgroup of 8 patients (including 4 who were studied during the first week) had a sleep study after 3 months of octreotide treatment to provide additional longitudinal data about the effect of octreotide on sleep apnea.

All patients were entered into a treatment protocol involving thrice daily administration (0800, 1400, and 2400 hours) of octreotide in total dosages of 100 µg, 200 µg, or 500 µg in a stepwise incremental design, as previously described [12]. Each patient was evaluated before and while receiving each dosage for growth hormone and IGF-1 suppression after an 8- to 12-week treatment period before progressing to the next higher dose. Patients studied during the first week of treatment received 100 µg of octreotide; during the third month the dosage was 200 µg, and during the sixth month it was 500 µg. The Stanford Sleepiness Scale [26] was administered during the morning to patients at baseline, 3 months, and 6 months to assess subjective response of daytime sleepiness to octreotide therapy. This is a widely used scale that has been validated against performance measures and consists of seven statements spanning gradations of subjective alertness from gradation 1 (wide awake) to gradation 7 (cannot stay awake).

Sleep Studies

Patients had overnight sleep studies in the Sleep Disorders Centre commencing at 2200 to 2300 hours and terminating at 0600 hours, as previously described [8]. We obtained the following indices from the sleep studies to assess the baseline severity of sleep apnea and the response to octreotide: respiratory disturbance index (RDI, the number of apneas and hypopneas [partial airway obstruction] per hour of sleep); the actual amount of sleep spent in apnea was also calculated (total apnea time) and separated into obstructive and central components (obstructive and central apnea time, respectively). Finally, the minimum oxygen saturation level reached during each apnea was measured and averaged (mean minimum oxygen saturation). Separate oxygen saturation values were calculated from non-rapid eye movement and rapid eye movement sleep.

Measurement of Endocrine Status

In 14 of 19 patients, growth hormone secretion was studied before and after 6 months of octreotide treatment from measurements obtained at hourly intervals during a 12-hour period. Random growth hormone measurements were obtained from the remaining 5 patients. The patients having monitoring at 3 months also had 12-hour growth hormone measurements, but those studied during the first week did not. Blood was sampled and growth hormone assays done, as previously described [8, 12]. Sleep studies were done separately at the Sleep Disorders Centre within 2 weeks of the growth hormone studies. All the samples from each patient's 12-hour study were assayed in one assay from which the mean growth hormone concentration was derived. Group growth hormone data analysis for the 19 patients included combined mean 12-hour measurements in the 14 patients and random growth hormone measurements from the 5 patients. The results were similar whether the combined data or only the mean 12-hour data were used. Levels of IGF-1 were measured from the fasting (first) sample obtained during a 12-hour growth hormone study [8, 12]. Biochemical remission was defined as an IGF-1 level of less than 35 nmol/L.

Statistical Analysis

Mean values are expressed ±SE. The effect of octreotide on sleep apnea was assessed by measuring RDI, mean minimum oxygen saturation during non-rapid eye movement sleep, mean minimum oxygen saturation during rapid eye movement sleep, total apnea time, and central and obstructive apnea time. For comparison of baseline with either 3- or 6-month data, paired t-testing was done. Confidence intervals (95% CIs) for the difference between baseline and 6 months were calculated. The Spearman rank correlation coefficient (rho) was used to further assess the relation between the biochemical and sleep apnea response to octreotide.

Results

Effect of Octreotide on Sleep Apnea

At baseline, various degrees of sleep apnea were present in the 19 patients (Table 1, Figure 1). Fourteen patients had severe sleep apnea (defined as more than 30 apneas/h or more than 20% of their sleep time spent in apnea, or both) with repeated decreases in oxygen saturation, impaired sleep quality characterized by increased amounts of light sleep (stages 1 to 2), and decreased amounts of slow wave (stage 3 to 4) sleep and rapid eye movement sleep. All 14 patients with severe sleep apnea reported marked sleepiness on the Stanford Sleepiness Scale with a mean score of 5 ±0.3. The remaining patients had only mild sleep apnea: Four patients had between 5 and 10 apneas per hour and 1 patient had an RDI of 2 apneas per hour. These patients reported minimal daytime sleepiness (Stanford Sleepiness Scores of either 1 or 2). Six of the 19 patients had predominantly central apnea (central apnea time > obstructive apnea time), and 2 other patients had frequent central apneas, although their predominant apnea type was obstructive.

View larger version:

• In this window
• In a new window

Figure 1. A decrease in the respiratory disturbance index (RDI) was observed in 13 of 14 patients with severe sleep apnea (RDI > 30 apneas/h). Data were measured at baseline ( = 19) and after 3 months ( = 8) and 6 months ( = 19) of octreotide therapy. Respiratory disturbance index in patients with acromegaly and sleep apnea before and during octreotide treatment.nnn

During the first week of octreotide treatment, the severity of sleep apnea did not change in the eight patients who were restudied in that time period. Although sleep apnea increased from 8 to 16 apneas per hour in one patient, the mean RDI for the group did not change (36 apneas/h compared with 33 apneas/h). Indices of oxygen saturation were not different when compared with pretreatment values. A 25% increase was noted in the proportion of sleep spent in rapid eye movement sleep (13.4% compared with 16.7%).

In the eight patients who were studied after 3 months of octreotide treatment, there was a decrease in RDI and total apnea time and a parallel increase in oxygen saturation during non-rapid eye movement and rapid eye movement sleep (Table 2, Figure 1). An increase was noted in the mean duration of rapid eye movement sleep.

After 6 months of octreotide, a 50% decrease was noted in the number of apneas and hypopneas per hour for the entire group with an associated improvement in the level of oxygen saturation in sleep (Table 1, Figure 1). Patients reported decreased snoring and subjective daytime sleepiness. An improvement was noted in sleep quality with an increase in the proportion of sleep spent in rapid eye movement sleep compared with baseline; a concomitant decrease in stage 2 sleep was also noted (Table 1). This improvement in sleep apnea and sleep quality was paralleled by better Stanford Sleepiness Scale scores (data not shown) indicating a decrease in sleepiness. In the 14 patients with severe sleep apnea, sleepiness scores decreased by 2 ±0.2 grades. This is equivalent to a change in reported state of alertness from “fogginess, beginning to lose interest in staying awake” to “relaxed, awake, and responsive, not at full alertness.”

The improvement in the amount of sleep spent in apneic breathing consisted of a decrease in obstructive and central components (Table 1, Figures 2 and 3). However, in the six patients with predominantly central apnea, the improvement in sleep apnea (decrease in the RDI from baseline compared with 6 months, 30 apneas/h compared with 14 apneas/h), was largely caused by a decrease in central apnea (baseline compared with 6 months, 12.2% of total sleep time compared with 4.1%) rather than any decrease in obstructive apnea (baseline compared with 6 months, 7.5% of total sleep time compared with 6.6%).

View larger version:

• In this window
• In a new window

Figure 2. Data were measured at baseline ( = 19) and after 3 months ( = 8) and 6 months ( = 19) of octreotide therapy. Obstructive apnea time expressed as percentage of total sleep time in patients with acromegaly and sleep apnea before and during octreotide treatment.nnn

View larger version:

• In this window
• In a new window

Figure 3. Five patients had no central sleep apnea recorded at either baseline or after 6 months of octreotide therapy. Data were measured at baseline ( = 19) and after 3 months ( = 8) and 6 months ( = 19) of octreotide therapy. Central apnea time expressed as a percentage of total sleep time in patients with acromegaly and sleep apnea before and during octreotide treatment.nnn

Effect of Octreotide on Growth Hormone and Insulin-like Growth Factor 1

Octreotide therapy decreased mean levels of growth hormone and IGF-1 (Tables 1 and 2, Figure 4). In the seven patients who had biochemical and sleep monitoring at baseline, 3 months, and 6 months, IGF-1 and growth hormone were decreased to a similar level at 3 months compared with 6 months (IGF-1, 52 nmol/L compared with 51 nmol/L; growth hormone, 7.6 µg/L compared with 6.0 µg/L). In 8 of the 19 patients, mean growth hormone and IGF-1 concentrations decreased to the normal range when assessed after 6 months of treatment (baseline compared with 6 months: IGF-1, 75 nmol/L compared with 29 nmol/L; growth hormone, 18.8 µg/L compared with 3.3 µg/L).

View larger version:

• In this window
• In a new window

Figure 4. Biochemical remission is defined as IGF-1 levels less than 35 nmol/L. There were 8 patients with remission and 11 patients without remission. Data are expressed as mean ±SE for baseline ( ) and after 6 months (solid bars) of octreotide therapy. Insulin-like growth factor 1 concentration and respiratory disturbance index in patients with and without biochemical remission of acromegaly after 6 months of octreotide treatment.open bars

Relation between Sleep Apnea and Biochemical Response to Therapy

In this analysis, we used mean growth hormone or IGF-1 levels as biochemical indices of disease activity and used RDI and mean minimum oxygen saturation as indices of apnea severity. At baseline, no relation was noted between mean levels of growth hormone or IGF-1 and RDI or mean minimum oxygen saturation. Although a decrease in mean levels of growth hormone and IGF-1 and an improvement in mean RDI and mean minimum oxygen saturation were seen in the group, no correlation was noted between the absolute decrease in growth hormone levels and the changes in RDI and mean minimum oxygen saturation (rho, −0.35; P > 0.2). The data were then analyzed to determine whether indices of apnea severity were different between the subgroup that attained biochemical remission (IGF-1 < 35 nmol/L) during octreotide treatment (8 patients) and the subgroup that did not (11 patients). Patients who attained biochemical remission had lower IGF-1 levels at baseline but similar RDI values (Figure 4). The mean RDI was similar for both subgroups after 6 months of octreotide treatment.

Discussion

We found that 6 months of treatment with octreotide (a somatostatin analog) improved sleep apnea in patients with acromegaly. By 3 months, central and obstructive apnea decreased in severity despite previous concerns [15-19]. We found no worsening of sleep apnea with either short- or long-term octreotide treatment. The relative proportion of time spent in rapid eye movement sleep increased with octreotide. This occurred before any detectable improvement in sleep apnea, suggesting a primary and independent effect of octreotide on sleep. The improvement in sleep apnea was accompanied by a decrease in IGF-1 and growth hormone, but no correlation was noted between the change in hormonal and sleep-breathing variables.

Few studies have systematically addressed whether treatment of acromegaly leads to an improvement in coexisting sleep apnea. Hart and colleagues [6] studied 21 patients with acromegaly and reported that only patients with active disease had sleep apnea [6]. For patients with acromegaly, we did not find any association between hormonal activity and the presence of sleep apnea, as previously reported [8]. Case reports have indicated marked improvement in the amount of sleep apnea after surgery [27], radiotherapy [28], or bromocriptine therapy [29]. However, Pekkarinen and coworkers [7] found no statistically different change in the amount of sleep apnea (measured by a static charge sensitive bed during a 2-hour daytime nap) despite decreased fasting growth hormone levels after hypophysectomy [7]. In contrast, our study, using detailed sleep and respiration monitoring throughout the night, as well as growth hormone measurements, found a decrease in sleep apnea and growth hormone hypersecretion after octreotide therapy. However, sleep apnea may either persist despite normalization of growth hormone levels or may improve markedly even if there is only partial biochemical remission.

In patients with acromegaly, soft-tissue swelling and macroglossia contribute to upper airway narrowing [30]. A decrease in the upper airway caliber increases the likelihood of upper airway obstruction in sleep [20]. Octreotide is reported to decrease soft-tissue swelling and tongue size in patients with acromegaly [9]. These changes would be likely to lead to improved upper airway dimensions and a decreased tendency for upper airway obstruction. This mechanism may possibly be responsible for the marked decrease in central apneas seen in this study. Central apnea often coexists with obstructive apnea in the same patient, and both forms of apnea respond to treatment aimed at preventing upper airway occlusion (for example, nasal continuous positive airway pressure [20, 31]). In some patients, central apnea may be secondary to reflex inhibition of respiration through activation of supraglottic mucosal receptors during oropharyngeal closure [31]. Support for this reflex mechanism is provided by the observation that upper airway anesthesia in patients with central apnea, including one patient with acromegaly, alters the breathing pattern in sleep from central to obstructive apnea [31].

Changes in respiratory control may be another explanation for the decrease in sleep apnea seen in association with octreotide therapy. Experimental models of human respiratory control have shown that abnormally high ventilatory responsiveness to chemical stimuli can induce central apnea and occasionally obstructive apnea [21-23]. Persons with increased ventilatory responsiveness to hypercapnia and hypoxia develop apnea in sleep [20, 22, 24]. Somatostatin decreases the ventilatory responses to acute and sustained hypoxia [18, 19] and, thus, may normalize breathing in persons with abnormally increased chemosensitivity. This may also explain why octreotide treatment did not lead to worsening of sleep apnea in this study, despite the profound apnea-promoting effects observed previously in rats [15-17]. It may also explain the improvement in central but not obstructive apnea observed in patients with predominantly central apnea. Nevertheless, prospective studies of breathing in sleep and simultaneous chemoreceptor activity and upper airway imaging are needed to clarify the mechanisms of apnea reduction in patients with acromegaly who receive octreotide therapy.

No close relation was noted between the degree of improvement in apnea severity and the decrease in biochemical activity. Maximal reduction in growth hormone and IGF-1 concentrations had occurred by the first 3 months of octreotide treatment, but sleep apnea severity continued to improve after 3 months. The patients who had biochemical remission (IGF-1 < 35 nmol/L) of acromegaly had a similar degree of sleep apnea after 6 months of octreotide treatment when compared with those with improved but still hormonally active disease (Figure 4). A discrepancy between the decline in biochemical activity in patients with acromegaly and other indices of clinical improvement has previously been observed during octreotide therapy [11]. There are several possible explanations for these findings: 1) Acromegaly may lead to variable or even permanent effects on the upper airway and, therefore, normalization of growth hormone and IGF-1 levels would not necessarily lead to normalization of upper airway anatomy; 2) improvements in sleep apnea during octreotide treatment may occur from direct effects on respiratory control or the upper airway, that is, independent of its action in decreasing growth hormone production by the pituitary; and 3) cointerventions, such as surgery or radiotherapy, may influence the relation between octreotide treatment and improvement in sleep apnea, although this is unlikely because only 7 of the 19 patients had previous treatment with these modalities and all 7 had persisting residual disease unresponsive to previous therapy.

We also found a marked increase in time spent in rapid eye movement sleep and a mild increase in slow wave sleep after 3 and 6 months of octreotide treatment. In patients with sleep apnea, the nadir oxygen desaturation and greatest apnea length occurs in rapid eye movement sleep [20], indicating that an increase in rapid eye movement sleep might lead to a worsening of sleep-disordered breathing. However, in this study, a decrease was noted in apnea severity. These changes in sleep pattern may be caused by decreases in growth hormone secretion [32] or decreases in sleep apnea [33]. The increase in rapid eye movement sleep occurred in the first week, even before any indications that sleep apnea had improved, suggesting a possible direct central effect of octreotide on sleep. Somatostatin and octreotide have been reported [34] to increase rapid eye movement sleep in rats. Regardless of the mechanism of the sleep pattern changes, the increase in rapid eye movement sleep and decrease in stages 1 and 2 sleep are likely to contribute to improved well-being, reduced tiredness and daytime sleepiness reported [9, 11] in patients with acromegaly after octreotide treatment. The improved sleep quality was reflected in the decreased daytime sleepiness noted on the Stanford Sleepiness Scale scores.

The decrease in apnea in these patients with acromegaly is probably secondary to octreotide therapy. Spontaneous improvement in patients with this severity of sleep apnea would be surprising in the absence of weight loss. Most studies show [35, 36] an increase in sleep apnea over time in patients with obstructive sleep apnea. However, despite improvement in sleep apnea severity, 10 of 14 patients with severe sleep apnea (RDI > 30 apneas/h) still met polysomnographic criteria for sleep apnea (RDI > 10 apneas/h) after 6 months of octreotide. Such patients need to be closely assessed and may require more definitive treatment for sleep apnea, such as nasal continuous positive airway pressure [37].

Abbreviations

IGF-1: insulin-like growth factor-1

RDI: respiratory disturbance index