Goldstein, Stephanie A.*; Hoffenberg, Edward J.†
Peutz-Jegher syndrome (PJS) is an autosomal dominant condition characterized by hamartomatous polyps throughout the gastrointestinal (GI) tract and mucocutaneous pigmentation (MP). The incidence of PJS is 1 in 8500 to 120,000 births (1,2). It is caused by a germline mutation in the STK11 gene (also known as LKB1) located on chromosome 19p13.3. It is associated with a serine threonine kinase that functions as a tumor suppressor (3,4), but the exact mechanism by which it produces tumors in PJS is unknown. A mutation or deletion is detected in about 80% of affected PJS families (5,6).
Polyp-related complications may develop in childhood. One third of children develop symptoms by age 10 years and half by 20 years (7). Polyp-related events include GI tract bleeding, anemia, small bowel obstruction, and intussusceptions, all of which may be life threatening. A recent large, prospective multicenter Dutch study by van Lier et al (8) reported cumulative intussusception risk in PJS of 15% by age 10 years and 50% by age 20. Patients with PJS also have a significantly increased risk of malignancy, primarily in adulthood, with an 81% to 94% lifetime cumulative risk for all cancers and an almost 70% risk for GI cancer (1,9,10).
Present guidelines for initiating screening for polyps vary. Some sources suggest starting at age 8 to 10 years with esophagogastroduodenoscopy (EGD) or capsule endoscopy only and if these are negative, colonoscopy at age 18 unless concerns develop (7,11). Similarly, interval follow-up recommendations vary from every year to every 5 years (7,11–13). These recommendations are derived from studies that are nonrandomized or descriptive (grade B) or expert opinion (grade C) (7) and focus on mitigating risk of cancer. Because malignancies rarely develop during childhood, these recommendations may not be relevant to monitoring in childhood.
At our institution, we recently encountered 2 young children with small bowel intussusception at age 5, one of whom had a large obstructing proximal small bowel polyp that also caused biochemical pancreatitis. We felt that it was appropriate to review our experience and see whether present guidelines were relevant to a pediatric population. We focused on the age at which symptoms appear, the age at which large polyps are first identified, and the age at which other clinical manifestations appear. We hoped to describe the rate at which new polyps develop in children and to look for similarities in the clinical course of related patients.
Medical records at Children's Hospital Colorado were reviewed from November 2000 to December 2011 for all of the children with diagnosis of PJS. Inclusion in the review required 2 or more of the following 3 criteria: 2 or more characteristic hamartomatous polyps of the small intestine, typical MP, and positive family history (FH) of PJS (7,10). The presence of a mutation in the STK11 gene was considered sufficient for inclusion in the present study.
Subjects were identified from a list of patients kept by one of the authors (E.J.H.) and then by using EPIC My Reports function with International Classification of Diseases-9 code 759.6. All of the patients who met the definition and who had undergone a clinical evaluation were included. The present study was approved by the Colorado Multiple Institutional Review Board.
Polyps were grouped by location: gastroduodenal, small bowel, or colon. Polyps were classified as “large” if >1 cm or “large” in the endoscopy, pathology, or radiology report.
The term “screening” was used for the initial evaluation, whether for diagnosis, investigation of a complaint, or initial search for polyps. “Surveillance” was used to indicate routine follow-up testing.
Initial Presentation and Screening
Fourteen children (10 boys) were identified; data at initial evaluation were not available for one. Median age at first clinical evaluation was 5 years (range 1–14 years). Demographic data, reasons for initial evaluation, and clinical course are summarized in Table 1. For 3 subjects, medical records from care at outside institutions were unavailable. Self-described ethnicity was Hispanic in 6; white in 5; and Asian, African American, or multiple ethnicities in 1 each.
Reasons for initial evaluation were: FH of PJS and/or MP in 8 (57%), rectal bleeding and/or prolapsed polyp in 4 (29%), intussusception in 1 (7%), gynecomastia in 1 (7%), and unknown in 2 (14%). Some had >1 reason for evaluation. All of the 14 screened subjects had MP.
The median age at first screening test (surgery, imaging, endoscopy, or capsule endoscopy) was 5 years (range 1–16). The median age at first identification of polyps was 8 years (range 1–18).
There were 6 subjects with initial clinic visit solely for FH/MP. Two subjects were first seen at age 4; one had intussusception and surgery at age 5 and the other had no clinically significant polyps at first endoscopy at age 14. One was first seen at age 7 and had a large antral polyp at age 7. Two subjects were first seen at age 8; 1 had intussusception and surgery at age 10, and 1 had a large pedunculated polyp in the stomach at first endoscopy at age 13.
Clinical Manifestations and Complications
During 11 years of the review, there were 7 intussusception events in 5 different subjects (2 events in 2 subjects and 1 event in 3 subjects) (Table 1). The median age at first intussusception was 10 years (range 5–16). Surgical reduction was required in 5 of the events. In 2 events, no polyp was identified as the lead-point (one at surgical exploration, the other through capsule endoscopy). Other complications included a prolapsed rectal polyp (PRP) in 2 children, both at age 2. SC tumor was identified in 2 of the 10 boys, at ages 8 and 11 years.
Sixty-six endoscopic procedures were performed (endoscopy, colonoscopy, and enteroscopy) with complications reported after 14 (22%) procedures. The only serious complications requiring admission were pancreatitis in 1 and bleeding requiring transfusion in 1. The remaining complications not requiring hospitalization or intervention were hematemesis (2), abdominal pain (2), and rectal bleeding (8). No capsule endoscopy complications occurred.
Single or multiple polyps were identified in 71 total procedures (endoscopy, capsule endoscopy, radiography). At initial evaluation, 9 subjects had polyps, with site identified as gastroduodenal in 5 (36%), small bowel in 7 (50%), and colon in 3 (21%). During follow-up, the involved areas were gastroduodenal in 12 (86%), small bowel in 11 (79%), and colon in 5 (36%) children. In 2 (14%), no polyps were identified at initial screening or during follow-up.
At least 1 large polyp was found in 9 of 14 (64%) children at median age of 7 years (range 1–16). Three children had only small polyps. Large polyps were fairly evenly distributed: 36% gastroduodenal, 36% small bowel, and 28% colon.
Of the 14 subjects, 10 were screened initially before age 8 years. Of these 10, 6 (43%) already had polyps and had follow-up evaluations a median of 13 months (range 11 months–7 years) later. Of the 4 patients with no polyp on initial screening, 1 has not yet had follow-up (age 4 at initial screen); 1 continued without polyps on surveillance after 1.5 years (at age 6); 1 had no polyps after 5 years at age 9, but had small bowel polyps after 10 years (at age 14); and 1 had gastroduodenal and colon polyps after 12 years (at age 18).
Phenotype Within Families
Two male siblings in family 1 are presently 8 and 11 years old. They were initially evaluated because of FH and both had gastroduodenal polyps at initial screening at 5 and 7 years of age. During follow-up, only the older sibling developed small bowel and colon polyps, an intussusception event, and SC tumor. The younger sibling remains free of polyps or complications to date.
A sister (13) and brother (11) in family 2 were evaluated because of FH and MP. Both have had multiple gastroduodenal and small bowel large polyps removed, the boy had SC tumor, and twice had intussusception with bowel resection.
The brother and sister twins in family 3 are now almost 5 years old. These twins were first evaluated at almost 3 years of age when both developed PRPs and blood in the stool. In follow-up, both have had gastroduodenal, small bowel, and colon polyps.
Family 4, a brother and sister now 20 and 17 years old, were initially screened at age 6 and 8 because of FH and MP. No polyps were noted on initial evaluations. Both have developed small gastroduodenal and small bowel polyps: the boy with multiple polyps at each site, the girl a single polyp at each site.
In our single institution experience, 40% of our 14 pediatric patients with PJS had already developed complications by age 8 years, including intussusception, rectal bleeding, PRP, and SC tumor. Published guidelines, however, recommend routine screening starting at age 8 or after.
We identified only 1 institution-specific pediatric experience PJS publication since 2000. This article, by Vidal et al from France (2), reports on 11 patients and with an experience similar to ours. The rate of complications before age 8 was 36% compared with our rate of 40%. The median age at first clinical indication of disease was 6 years (range 4–15), similar to the median age of 5 for first polyp identification in our study. In contrast, a retrospective study of adults with PJS suggests an older age of 13 for first manifestation or polypectomy (14). None of the 3 recent guidelines recommends routine endoscopic screening before age 8 years (1,7,11). Based on our study and that of Vidal et al (2), it appears that screening at age younger than 8 years, before the first complications develop, would be appropriate.
The reasons for first evaluation in childhood PJS have not been reported previously. Because an FH is not present in about 25% of cases, which represent cases of de novo mutation, FH alone will not suffice to identify all of the cases (15). In our series, the 2 most common reasons for initial evaluation were FH (47%) and MP (47%); blood in the stool (29%) was much less common.
Only 1 patient in our series presented initially with intussusception. In a large Dutch series of 110 patients with PJS, 69% suffered at least 1 intussusception with median age of 16 (range 3–50) for first event (8). At age 10, the estimated cumulative intussusception risk was 15% (95% CI 8–22). In the St Mark's Registry, of 34 English adults with PJS, 68% reported laparotomy for obstruction before the age of 18 (16). Because intussusception or obstruction is common in childhood, early evaluation may prevent these complications.
The recent guidelines on when and how to perform initial screening vary. A Dutch surveillance recommendation suggests starting at age 10 with capsule endoscopy, and if no polyps are identified, continuing with capsule endoscopy every 2 to 3 years (1). This recommendation moves screening earlier, but, unless polyps are identified on capsule endoscopy, delays endoscopy, and colonoscopy until age 20. Subsequently, the same Dutch group reported a significant intussusception risk of 15% by age 10 (8). By combining our data with those of Vidal et al (2), one-quarter of the combined 25 children experienced intussusception before age 10. Thus, there is accumulating evidence that earlier screening and intervention may be warranted; however, the effectiveness of earlier screening in reducing morbidity such as intussusception, obstruction, or bleeding remains unproven.
In addition to intussusception risk, factors such as polyp distribution and size may be important in determining how children with PJS should be screened and then followed. Gastroduodenal polyps seem more common in early childhood than in adults. Gastroduodenal polyps were identified in 73% (our study) and 60% (Vidal et al) (2) of children, a rate higher than the 25% reported in adults (10). In contrast, the frequency of colonic polyps may be similar in children and adults: 33% (our study) and 50% (2) in children is comparable with 30% in adults (10). Because polyps outside of the small bowel may cause symptoms, EGD and colonoscopy plus capsule endoscopy may be more accurate than capsule endoscopy alone in identifying polyps causing morbidity in childhood.
Polyp size considerations may also need modification for the small child. The present recommendation is to remove polyps >15 mm (8); however, because children have smaller lumen size than adults, smaller polyps may potentially cause obstruction or intussusception. We could find no data comparing size of polyps causing childhood or adult intussusception. In children, pedunculated polyps, regardless of size, should be considered for removal. This approach of cleaning out all polyps has been described as a “clean sweep,” but there are no studies to show effectiveness in reducing surgeries and complications (17).
Based on the early age of complications in early childhood, we recommend that screening should aim to evaluate the entire intestinal tract, and begin around 4 to 5 years of age with capsule endoscopy, EGD, and colonoscopy as well as for SC tumors in boys with feminizing features and girls with precocious puberty (1,7,11,18,19) (Table 2).
More aggressive monitoring strategies must be weighed against cost and risk. Having a polypectomy confers a higher risk of bleeding or perforation than colonoscopy alone (20), but our data of no perforations and only 1 significant bleed in 66 procedures suggest that this risk can be low.
In evaluating the child with high index of suspicion for PJS because of FH, SC tumor, or MP, consideration should be given to evaluation of the entire GI tract for polyps even when genetic testing is negative. We did not detect any clear genotype-phenotype correlation in the 4 sibling pairs. Truncating mutations may be associated with a more severe course (21), whereas missense mutations may be associated with later development of polyps (14); however, genotype correlation with cancer risk or intussusception risk has not been shown (8,22,23). At this time, guidelines do not recommend genetic testing nor do the results of genetic testing affect screening or follow-up.
Selection bias is a significant limitation of the present study. Because this is a single-center retrospective chart review, and not a population-based study, asymptomatic children may not have been seen, and this bias may inflate the rate of early complications. The size and distribution of polyps was not assessed in a standard fashion because of the retrospective nature of the present study. Polyp size could only be designated as small or large based on the description in medical record. Because there was no standard protocol for follow-up of PJS in our institution, we were not able to assess rate of new polyp formation (or of missed polyps) or the effectiveness of surveillance strategies in childhood.
Evaluation for and management of small bowel polyps continue to evolve. New nonradiating imaging techniques such as video capsule endoscopy (24), magnetic resonance enterography (25), and new interventional techniques such as single- and double-balloon enteroscopy (26) show promise for the prevention of morbidity, but are still unproven.
Our review confirms that PJS polyps occur at an early age and in all areas of the GI tract in children with PJS and young children do present with consequences of these polyps. Our data build additional evidence to suggest that present recommendations may need revision to investigate the entire GI tract earlier in childhood. Our experience has led to a modification of our approach to children with PJS. We now begin initial screening at age 4 to 5 years with capsule endoscopy, EGD, and colonoscopy and evaluation of boys for SC tumors. We perform an annual examination and, once polyps are removed, recommend surveillance every 2 to 3 years with the same 3 procedures. Repeat evaluation is indicated upon development of symptoms suggestive of polyp-associated complications. Monitoring this population into adulthood may allow conclusions as to whether this more aggressive approach will indeed prevent complications.
The authors thank Judith M. Sondheimer, MD, for writing assistance.
1. van Lier MG, Wagner A, Mathus-Vliegen EM, et al. High cancer risk in Peutz-Jeghers syndrome: a systematic review and surveillance recommendations. Am J Gastroenterol
2. Vidal I, Podevin G, Piloquet H, et al. Follow-up and surgical management of Peutz-Jeghers syndrome in children. J Pediatr Gastroenterol Nutr
3. Hemminki A, Markie D, Tomlinson I, et al. A serine/threonine kinase gene defective in Peutz-Jeghers syndrome. Nature
4. Jenne DE, Reimann H, Nezu J, et al. Peutz-Jeghers syndrome is caused by mutations in a novel serine threonine kinase. Nat Genet
5. Hezel AF, Bardeesy N. LKB1; linking cell structure and tumor suppression. Oncogene
6. Volikos E, Robinson J, Aittomaki K, et al. LKB1 exonic and whole gene deletions are a common cause of Peutz-Jeghers syndrome. J Med Genet
7. Beggs AD, Latchford AR, Vasen HF, et al. Peutz-Jeghers syndrome: a systematic review and recommendations for management. Gut
8. van Lier MG, Mathus-Vliegen EM, Wagner A, et al. High cumulative risk of intussusception in patients with Peutz-Jeghers syndrome: time to update surveillance guidelines? Am J Gastroenterol
9. Lynch HT, Lynch JF, Lynch PM, et al. Hereditary colorectal cancer syndromes: molecular genetics, genetic counseling, diagnosis and management. Fam Cancer
10. Jasperson KW, Tuohy TM, Neklason DW, et al. Hereditary and familial colon cancer. Gastroenterology
11. Giardiello FM, Trimbath JD. Peutz-Jeghers syndrome and management recommendations. Clin Gastroenterol Hepatol
12. Dunlop MG. Guidance on gastrointestinal surveillance for hereditary non-polyposis colorectal cancer, familial adenomatous polypolis, juvenile polyposis, and Peutz-Jeghers syndrome. Gut
2002; 51 (Suppl 5):V21–V27.
13. Hemminki A. The molecular basis and clinical aspects of Peutz-Jeghers syndrome. Cell Mol Life Sci
14. Amos CI, Keitheri-Cheteri MB, Sabripour M, et al. Genotype-phenotype correlations in Peutz-Jeghers syndrome. J Med Genet
15. Gammon A, Jasperson K, Kohlmann W, et al. Hamartomatous polyposis syndromes. Best Pract Res Clin Gastroenterol
16. Hinds R, Philp C, Hyer W, et al. Complications of childhood Peutz-Jeghers syndrome: implications for pediatric screening. J Pediatr Gastroenterol Nutr
17. Oncel M, Remzi FH, Church JM, et al. Benefits of ’clean sweep’ in Peutz-Jeghers patients. Colorectal Dis
18. Howell L, Bader A, Mullassery D, et al. Sertoli Leydig cell ovarian tumour and gastric polyps as presenting features of Peutz-Jeghers syndrome. Pediatr Blood Cancer
19. Massa G, Roggen N, Renard M, et al. Germline mutation in the STK11 gene in a girl with an ovarian Sertoli cell tumour. Eur J Pediatr
20. Rabeneck L, Paszat LF, Hilsden RJ, et al. Bleeding and perforation after outpatient colonoscopy and their risk factors in usual clinical practice. Gastroenterology
21. Salloch H, Reinacher-Schick A, Schulmann K, et al. Truncating mutations in Peutz-Jeghers syndrome are associated with more polyps, surgical interventions and cancers. Int J Colorectal Dis
22. Hearle N, Schumacher V, Menko FH, et al. Frequency and spectrum of cancers in the Peutz-Jeghers syndrome. Clin Cancer Res
23. Hearle N, Schumacher V, Menko FH, et al. STK11 status and intussusception risk in Peutz-Jeghers syndrome. J Med Genet
24. Postgate A, Hyer W, Phillips R, et al. Feasibility of video capsule endoscopy in the management of children with Peutz-Jeghers syndrome: a blinded comparison with barium enterography for the detection of small bowel polyps. J Pediatr Gastroenterol Nutr
25. Caspari R, von Falkenhausen M, Krautmacher C, et al. Comparison of capsule endoscopy and magnetic resonance imaging for the detection of polyps of the small intestine in patients with familial adenomatous polyposis or with Peutz-Jeghers’ syndrome. Endoscopy
26. Hsu TC, Lee TC, Chiu HM, et al. Radical endoscopic polypectomy combined with double-balloon enteroscopy and colonoscopy for Peutz-Jeghers syndrome. J Pediatr Gastroenterol Nutr