AIM: To investigate the long-term follow-up of distal intestinal obstruction syndrome (DIOS) in Israeli cystic fibrosis (CF) patients.

METHODS: This is a multi-center, comparative, retrospective study in which we reviewed the medical records of all CF patients from three major CF centers in Israel who were treated in the period from 1980 to 2012. Patients diagnosed with DIOS were defined as the study group. The patients were diagnosed with DIOS based on their clinical presentation and typical findings on either abdominal X-ray or computerized tomography scan. For the control group, CF patients with no DIOS were matched to the patients in the study group for age, sex, and cystic fibrosis transmembrane conductance regulator (CFTR) mutations. For both groups, the collected data included age, sex, CFTR genotype, weight, height, and body mass index. Clinical data included respiratory function tests in the last five years prior to the study, respiratory function test immediately before and after the DIOS event, number of hospitalizations, sputum culture results, and CF-related conditions diagnosed according to the CF clinical practice guidelines. In the study group, data on the DIOS treatment and tendency for DIOS recurrence were also analyzed.

RESULTS: The medical charts for a total of 350 CF patients were reviewed. Of the 350 CF patients, 26 (7.4%) were diagnosed with DIOS. The control group included 31 CF patients with no DIOS diagnosis. The mean follow-up period was 21.6 ± 8.2 years. The total of DIOS episodes in the follow-up period was 60. The distribution of DIOS episodes was as follows: 6/26 (23.1%) study patients had one episode of DIOS in their lifetime, 7/26 (26.9%) had two episodes, 7/26 (26.9%) had three episodes, and 6/26 (23.1%) had four or more episodes. Compared to the control group, DIOS patients had a significantly higher incidence of meconium ileus in the past (65.4% vs 0%, respectively, P < 0.02), more Aspergillus spp. colonization (34.6% vs 3.2%, respectively, P < 0.02), and a higher number of hospitalizations due to respiratory exacerbations (8.6 vs 6.2 mean total hospitalizations per follow-up period, respectively, P < 0.02). No other significant differences were found between the control and study groups. The conservative treatment of DIOS, which mainly includes hydration and stool softeners, was successful in 82% of the episodes. The survival rate was similar for both groups.

CONCLUSION: CF patients with DIOS suffer from recurrent hospitalizations and airway pathogen acquisition. Although recurrence of DIOS is common, conservative treatment is successful in most patients.

Cystic fibrosis (CF) is an autosomal recessive disease caused by mutation of the cystic fibrosis transmembrane conductance regulator (CFTR) protein, which is a cell membrane channel for transport of chloride ions[1]. The systems usually affected by this disease are respiratory, gastrointestinal, and reproductive[2]. The prevalence of CF-related gastrointestinal complications has been on the rise in recent years, a fact that can partly be attributed to an increase in the average age of CF patients[3]. The two gastrointestinal manifestations in CF that lead to intestinal obstruction are meconium ileus (MI) and distal intestinal obstruction syndrome (DIOS). MI is a condition specific to infants, who in most cases are also diagnosed with CF, strongly suggesting that MI may be the first clinical marker of the disease. The MI appears in 10%-21% of CF patients[4-6]. The meconium in the MI neonates is viscous, dry and thick, consequently causing intestinal obstruction. DIOS, a similar phenomenon of intestinal obstruction, occurs later in life. DIOS is characterized by the accumulation of viscous intestinal contents within the bowel lumen involving the terminal ileum, cecum and ascending colon, potentially causing complete or partial intestinal obstruction[7]. The DIOS incidence rate ranges from 2.3% to 41.3%[8]. A recent study in European children showed complete obstruction incidence rate ranging 5-12 episodes per 1000 CF patients per year[9].

Some of the indications of the presence of acute complete obstruction are bilious vomiting, a mass in the ileocecal area, abdominal pain and distention, and air-fluid level in the small intestine on abdominal X-ray. The incomplete fecal obstruction is characterized by abdominal pain and a mass in the ileocecal area[9]. Although the pathogenesis of DIOS is unknown, the underlying assumption is that DIOS is a combination of a few factors, including pancreatic insufficiency (PI) with abnormal secretion of pancreatic enzymes and abnormal digestion, insufficient secretion of bicarbonate, viscous and mucoid secretions of the intestinal mucosa, and impaired intestinal motility[9]. Various DIOS treatment options are available[3,9]. In most cases, patients with incomplete DIOS respond well to oral rehydration combined with stool softeners or laxatives. When DIOS presents with more severe intestinal obstruction characterized by bilious vomiting or when washout therapy has failed, hospitalization is recommended, and intravenous rehydration and nasogastric aspiration should be commenced. With early aggressive medical management, surgery is seldom required; laparotomy with washout via enterostomy should supersede resection of the ileocecum.

There is a lack of consensus in the literature on the different aspects of DIOS in CF patients. Correlation between DIOS and the pulmonary and nutritional status was found to be either non-significant[7] or positive[3]. On the other hand, DIOS patients suffered from a lower forced expired volume in 1 s (FEV1) and had more severe morbidity than patients without DIOS[9,10]. The correlation between pancreatic enzyme replacement therapy (PERT) and DIOS is still not fully understood. Some reports suggest a positive correlation between PERT and an increased risk of obstruction[9,11]. Others found that the pancreatic failure and uncontrolled steatorrhea led to an increase in intestinal obstruction frequency that did not improve even after the introduction of a more efficient pancreatic enzyme treatment[3,10,12-14]. The correlation between DIOS and MI also remains inconclusive, ranging from positive[2,8,9] to non-significant[10]. Contradictory evidence exists for the correlation between severity of the CFTR mutation and the chance of developing DIOS, as well[8-10]. Additional DIOS risk factors include dehydration, CF-related diabetes, and organ transplantation[15].

The objective of the current study was to assess clinical characteristics, risk factors, morbidity and mortality among Israeli CF patients with and without DIOS during a 20-year follow-up period. Furthermore, the treatments used by the three CF centers for DIOS-positive CF patients were also assessed. A more comprehensive understanding of DIOS and a development of a clear treatment strategy will help create an all-encompassing management plan to be used by patients and doctors to manage this complicated disease.

The study encompassed 350 patients, accounting for about 60% of all CF patients in Israel. Of the 350 patients, 26 (7.4%) were diagnosed with DIOS. The average length of the follow-up period for the entire cohort was 21.6 ± 8.2 years. Twenty (76.9%) DIOS patients and 26 (83.8%) control patients underwent follow-up for > 15 years, and 17 (65.3%) DIOS patients and 19 (61.3%) control patients were followed for ≥ 20 years (Table 1). The average age for occurrence of the first DIOS episode was 9.7 ± 8.5-years-old. Following this first episode, 13 patients (50.0%) were followed for at least 10 years and 6 (23.1%) patients were followed for at least 15 years. The average follow-up period after the first DIOS episode was 9.5 ± 7.7 years. Six (23.1%) study patients had one episode of DIOS in their lifetime, 7 (26.9%) had two episodes, 7 (26.9%) had three episodes, and 6 (23.1%) had four or more episodes. The control group included 31 CF patients not diagnosed with DIOS matched by age, sex, and CFTR genotype to the study group. The clinical characteristics of the 57 patients included in the study are presented in Table 2.

From the DIOS group, 13 (50.0%) patients were homozygous and 8 patients (30.7%) were compound heterozygous for the CFTR gene. In the control group, 12 (38.7%) patients were homozygous and 9 (35.4%) patients were compound heterozygous for CFTR gene. The rest of the patients in both groups had unknown mutations. The two prominent alleles were DeltaF508 and W1282X. DeltaF508 was found in 46.2% of the DIOS patients and 58.1% of the control group patients. W1282X was found only in 46.2% of the DIOS group patients. Other mutations included: T360K, 3849+10KBC→T, G85E, D952H, R334W, 405+1G→A, S549R, 5T, Q359K, N1303K, 3120del, e92x, and I1234V. There was no statistical difference between the DIOS and control group in the severity of mutations or mutation class. In the DIOS group, the BMI was slightly lower than that in the control group but the difference was not significant (18.4 vs 19.3, respectively, P = 0.301). The percentage of patients who received percutaneous endoscopic gastrostomy tube insertion, to improve nutritional status, was similar between the DIOS group and the control group (7.7% vs 6.5%, respectively, P = 1.00). In the DIOS group, 18 patients had pancreatic insufficiency and were treated with an enzyme replacement therapy. There was no statistical difference in PI between the DIOS group and the control group [18 (69.2%) vs 27 (87.1%), respectively, P = 0.97]. The incidence of MI in the medical history of DIOS group patients compared to the control group was 65.4% vs 0.0%, respectively (P = 0.001).

The incidence rate and diversity of bacterial colonization by Pseudomonas aeruginosa, Staphylococcus aureus, Burkholderia cepacia, and nontuberculous Mycobacterium was similar between the DIOS and control groups (84.6% vs 90.3%, P = 0.691; 53.8% vs 41.9%, P = 0.431; 7.7% vs 0%, P = 0.204; and 7.7% vs 16.1%, P = 0.436, respectively). In contrast, patients suffering from DIOS had significantly more infections with Aspergillus spp. then the control group (34.6% vs 3.2%, P = 0.003, respectively).

The patients who developed intestinal obstruction did not suffer from an increase in the number of CF-related complications (42.3% vs 58.1% in the controls, P = 0.292). There was an upward trend in the number of organ transplantations in the DIOS group (15.4% vs 3.2%, respectively, P = 0.167) and an increase in rectal prolapse in the control group (0.0% vs 6.5%, respectively, P = 0.0495).

In order to evaluate the severity of lung disease, pulmonary function tests (PFTs) performed during the 5-year period prior to the study (2007 - 2011) were reviewed. The average value of each PFT parameter was calculated from all the yearly PFTs. There were no significant differences between the DIOS and control groups, as seen in Table 3. Further evaluation of the effect of an intestinal obstruction episode on lung disease included comparison of the PFTs of DIOS patients collected up to 2 mo before, and PTFs collected 2 to 4 wk after an obstruction. There was no significant difference between PFTs in the period prior to or following the bowel obstruction. The average FEV1 was 68.6% before vs 70.2% after obstruction (P = 0.979), the average forced vital capacity was 76.35% before vs 76.4% after obstruction (P = 0.759), and the average forced expiratory flow between 25% and 75% of forced vital capacity was 54.9% before vs 50.0% after obstruction (P = 0.418). The patients in the study group had significantly more hospitalizations due to respiratory exacerbations than those in the control group (8.6 vs 6.2 mean total hospitalizations per follow-up period, respectively, P = 0.018). In order to exclude age bias, the average number of hospitalizations per year for a 20-year follow-up period was measured. Even with this correction, our results still show a significantly higher number of hospitalizations in the DIOS group compared to the control group (0.4 vs 0.3 mean total hospitalizations per year, respectively, P = 0.008).

Overall, the DIOS group included 26 patients who suffered 60 episodes of intestinal obstruction during the 20-year follow-up period. The treatment distribution for the 60 episodes was as follows: 75.0% of episodes were managed conservatively, 18.3% surgically, and 6.7% resolved without any intervention. The most often used drug for the treatment of intestinal obstruction was oral administration of Gastrografin, and the least used drug was N-acetylcysteine (Table 4). As part of supportive treatment, 61.6% of the DIOS patients received antibiotics and fluids. There was one mortality case in each group. In the study group, death was due to complications related to liver transplantation, not associated with DIOS; the patient in the control group died due to the end-stage lung disease.

This study followed Israeli CF patients who suffered from DIOS for an average follow-up period of 21.6 ± 8.2 years after the first bout of bowel obstruction. In the 60% of all Israeli CF patients, the DIOS incidence rate was 7.4%. This rate most likely represents the actual rate of the entire CF patient population in Israel, and it falls within the wide range of DIOS incidence worldwide of 2.4%-41.3%[8]. This incidence can be due to mutations, the hot climate, and/or food that are specific to Israel.

In our study, none of the patients in the control group had a prior history of MI, while 65.4% of the patients in the study group did have it (P = 0.001). The intestinal obstruction in the neonate, which is frequently managed surgically, leaves a diseased intestine, increasing the possibility for future adhesions and subsequent obstructions. MI may be the primary indication of intestinal susceptibility that makes the intestine prone to recurrent obstructions. Of the 26 DIOS patients, 20 (76.9%) had more than one episode of intestinal obstruction in their lifetime, supporting these two possible explanations. Several groups reported a significant correlation between DIOS and a history of MI[8]. O’Halloran et al[18] showed MI incidence of 28% in patients with DIOS. Houwen et al[9] found an even higher incidence of MI in DIOS patients (44%). A significant positive correlation was also supported by the findings from a study by Efrati et al[2] that used the same unique CF population in Israel as our study. Since both MI and DIOS probably share similar pathophysiology, this association seems logical. However, Dray et al[10] found no significant correlation between DIOS and a history of MI. In our study, DIOS patients had a significantly higher hospitalization rate mostly due to respiratory exacerbations, although some of the hospitalizations were due to intestinal obstructions.

Despite the increased number of hospitalizations, the results of our study show that morbidity and mortality rates of the DIOS patients were similar to those of the control group. This finding is surprising since one would expect every hospitalization to be related to the patient’s baseline deteriorating. Similar results have been shown by Koletzko et al[7] who found no correlation between DIOS and severity of lung disease, with the respiratory function of DIOS patients being similar to CF patients without DIOS in terms of the severity spectrum. Khoshoo et al[3] showed that DIOS is related to a less severe pulmonary disease. It appears that the administration of the immediate treatment to resolve the acute DIOS event, as well as the long-term supportive care given to these patients, contributed to these good results. Treatments that demonstrated beneficial properties in decreasing the damage and increasing recovery of the patients to their pre-hospitalization state include antibiotic treatment, Gastrografin, long-term laxative treatment, stool softeners, fluids, nutritional support, and chest physiotherapy. Still, other studies have found an association between DIOS and higher morbidity[9,10].

The survival rate was similar for the two groups in this study, suggesting that DIOS does not increase the mortality rate of CF patients. In addition, as shown previously, there are similarities between the DIOS patients and the control regarding the respiratory function tests and nutritional status. An interesting result is that the study patients were not sicker in the long run and did not deteriorate more than the control patients, despite their need for more treatments and recurrent hospitalizations. A potential explanation of this observation is that the majority of DIOS patients suffered from MI as neonates and were diagnosed early as CF patients, therefore receiving close medical attention and aggressive treatment very early in life. This early diagnosis gave them an advantage by preventing or postponing irreversible lung damage. Indeed, several studies have shown that an early diagnosis of CF improves prognosis in the long run[2,19,20]. Moreover, Efrati et al[2] showed that patients with MI showed no differences regarding PFTs in comparison to patients who did not suffer from MI after a 25-year follow-up.

CF patients suffer from recurrent respiratory tract infections, in part due to dehydration of the airway surface and decreased mucociliary clearance which may lead to chronic lung infections[21]. As a result, various bacterial pathogens such as Pseudomonas aeruginosa or Staphylococcus aureus colonize the respiratory airways in all of the CF patients regardless of DIOS diagnosis. However, Aspergillus spp. colonization is seen more frequently in patients who suffer from DIOS. We hypothesize that frequent hospitalizations expose DIOS patients to a higher number of antibiotic treatments, creating an optimal microenvironment for fungal infections.

The majority of CF patients, who participated in our study, suffered from pancreatic insufficiency and their condition was managed with PERT. Furthermore, there was no statistical difference in the PI prevalence and use of PERT between the DIOS group and the control group. These findings contradict previous studies that showed a connection between the aggressive use of PERT during the years and an increased risk of DIOS[9,22]. Others have shown an increase in intestinal obstruction and the DIOS incidence due to pancreatic failure and uncontrolled steatorrhea despite the introduction of more efficient pancreatic enzyme treatment[3,10,13-15].

The current study was not able to establish a clear link between CFTR mutations and DIOS as demonstrated in prior studies. The Cystic Fibrosis Genotype-Phenotype Consortium[23] and others[24-26] showed a strong association between severe mutations and pancreatic insufficiency. DIOS is mainly seen in patients with severe CFTR-related phenotypes[8,10]. Houwen et al[9] showed that 82% of DIOS patients had severe CFTR mutations vs 3% who had non-severe mutations. However, even patients with a mild phenotype developed DIOS, which suggests that not only CFTR but other modifier genes may have an effect on the severity of the gastrointestinal phenotype[9]. The results from a study by Blackman et al[8] did not support this statement and the authors argued that modifier genes do not contribute to DIOS, as opposed to MI.

In our study, the two most-prominent mutations in the patients suffering from DIOS were DeltaF508 and W1282X. Among the study patients, 46.2% carried at least one DeltaF508 allele, and 36.5% of the alleles of the entire cohort carried this mutation. Prior studies reported this mutation as the most-common within the CF patient population worldwide, being found in 67% of chromosomes[23]. In addition, the correlation between DeltaF508 and pancreatic insufficiency has previously been reported[23]. Indeed, 45 (78.9%) patients from the entire cohort suffered from PI. The second most-prominent mutation in the study group was W1282X, found in 46.2% of our DIOS patients in at least one allele. This mutation is almost exclusively found in the Jewish Ashkenazi population, which constitutes 1% of CF patients in the world[27]. Although morbidity and mortality of the W1282X mutation have not been studied in depth, the mutation is considered to be a Class I stop mutation. Our study found that the respiratory, nutritional and morbidity status, as well as the mortality of the study patients, were all similar to that of the control group. The unique W1282X mutation, more prevalent in our patient population, compared to the world CF population, may be contributing to this fact. It is unlikely that this severe mutation would have a protective effect on the patients. Most likely, the W1281X mutation enables modifier genes to affect the clinical presentation of DIOS, consequently providing a protective effect. In light of our study results, one can concede that modifier genes or the mutation itself may indeed be a factor affecting the clinical presentation of DIOS.

In this study we demonstrated the efficacy of supportive and conservative management of intestinal obstruction with rare need for surgical treatment. Oral Gastrografin was the most commonly used treatment. Indeed, of the 45 episodes managed in a conservative manner, 15 (33.3%) were resolved solely by Gastrografin and 9 (20%) were resolved by Gastrografin with an additional oral agent or enema as part of a combined therapy. Combined therapy was initiated either as the first choice or when Gastrografin alone did not resolve the obstruction. If the combined therapy was initiated as a second line after the failure of Gastrografin to resolve the obstruction, then the success rate of the Gastrografin treatment was 57.7%. However, since combined therapy was initiated as the first choice in some cases, usually due to severe obstruction, then the actual success rate of Gastrografin was even higher. In the past, N-acetylcysteine treatment either orally or via enema, was highly accepted[7,28-30]. However, some investigators have shown that treatment with N-acetylcysteine either has a high failure rate[7] or that a high dose increases the risk of liver damage[7,31]. Furthermore, N-acetylcysteine is cost-prohibitive and due to offensive taste patients often fail to comply with the treatment[32]. In comparison, Gastrografin, a hyperosmolar dye (1500-2000 mOsm/kg), has a success rate of up to 81% with a single oral dose, and patients prefer this treatment over other available treatments[18]. However, the potential risks associated with taking Gastrografin orally are vomiting or coughing while drinking, resulting in aspiration of the drug[7], causing pulmonary edema and acute inflammation with alveolar damage and bleeding, due to the high osmolality of the drug[6,7]. Houwen et al[9] related the use of Gastrografin to enema. The group emphasized the importance of the substance as both a diagnostic and therapeutic tool (radiopaque and hyperosmolar) followed with caution regarding the danger of translocation of large quantities of water from the circulation to the intestine. Their conclusion was that this procedure should be administrated to older children with appropriate monitoring[9]. Our study refutes these concerns regarding Gastrografin. Indeed, being the most-common treatment, the study patients did not have lower respiratory function tests in comparison to the control, the mentioned complications were not observed during the treatment, and the treatment had a high success rate for resolving the obstruction. In contrast, the previously popular treatment involving the use of enema with N-acetylcysteine was rarely used. Surgical treatment, which was the only definitive treatment in the past, is associated with high morbidity and mortality and was rarely used and required in the current study. The high morbidity and mortality rates associated with surgical treatment are related to bleeding, delayed healing of surgical wounds, and postoperative infections[29]. Therefore, as of recently the application of surgical treatment has been decreasing and in most cases conservative treatment is the primary choice. However, researchers like Speck et al[33] suggest that with an increase in the incidence of DIOS in the adult CF population, the need for surgical intervention will rise; but this did not seem to be the case in our study population.

Our study has a few limitations. Primarily, the data were taken from three different centers, each using different follow-up and report systems. On the other hand, these three centers are all located in Israel; therefore, the treatment strategies were not significantly different in terms of drug choice and treatment regimen. A prospective study would be the best approach for determining the efficacy of different treatment strategies in the management of intestinal obstruction. However, our study is retrospective and was based on a review of the treatments given through the years to treat different episodes and could, therefore, be affected by different habits or tendencies of centers to offer one agent over another as a primary treatment option. Nonetheless, oral Gastrografin was the most-common agent used by all three centers and showed very high success rates in each. Because Gastrografin has few side effects and forms of administration, we strongly recommend this compound as the primary choice for treating intestinal obstruction in CF patients diagnosed with DIOS. Despite the above-mentioned limitations, we believe that these results can be implemented for the entire CF patient population worldwide.

In conclusion, CF patients, both pediatric and adult, who suffer from recurrent intestinal obstruction either in the form of MI or DIOS, have similar pulmonary and nutritional status, survival rates and associated morbidities compared to CF patients who do not. This finding is encouraging considering that DIOS patients often suffer from high rates of hospitalization and respiratory tract infections. Conservative treatment, especially with oral Gastrografin, is very successful in these patients. The combination of early diagnosis, appropriate treatment, and close follow-up all lead to successful results in CF patients of older age. The more we expand the knowledge regarding the treatment of DIOS in CF patients, the more efficient the clinical management approach will become, potentially significantly improving the life expectancy and the quality of life for these patients.