Long-Term Outcomes of Paediatric-Onset Craniopharyngioma: A Retrospective Analysis from a Tertiary Care Centre in North India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.344661
Source of Support: None, Conflict of Interest: None
Keywords: Craniopharyngioma, craniopharyngioma recurrence, long-term morbidities, pediatric-onset, pituitary hormone deficits, short-stature, visual defects
Craniopharyngioma are rare intracranial neoplasms with an incidence of 0.5–2 cases per million person-years, of which 30%–50% are apparent during childhood and adolescence. Due to its location in the sellar-suprasellar region, pituitary hormone deficiencies, visual deficits, and various hypothalamic dysfunctions are common morbidities associated with craniopharyngioma both at the presentation and during the long-term follow-up post-treatment. Recurrence and progression of these tumors is a significant morbidity in itself. Combined surgical and radiation treatment results in lower recurrence rates than surgery alone.
These morbidities cause serious impairment of quality of life. Timely detection and management of these morbidities through regular long-term follow-up is an integral part of the comprehensive management of craniopharyngioma.
There are no studies of childhood-onset craniopharyngioma from India addressing the long-term clinical outcomes and associated morbidities.
Setting: Children ≤18 years of age at presentation who were diagnosed and treated at a tertiary care referral hospital and institute in Uttar Pradesh, India. The hospital is a referral center for the states of Uttar Pradesh and adjoining Bihar and Madhya Pradesh. Ethical approval was obtained from the institute's ethics committee with a waiver of consent.
Subjects: All children aged ≤18 years at presentation diagnosed with histopathology confirmed craniopharyngioma between 2003 and 2018 with at least 12 months of follow-up period were included.
Children with craniopharyngioma are followed at 3–6 monthly intervals in the endocrinology outpatient department of our hospital.
Data collection: The details were obtained from the hospital information system records and patients' file records. Patients were also telephonically contacted to get information as needed and when possible. Following are the details of various data collection methods:
Statistical analysis: All continuous variables are expressed as mean ± SD or median (range) as appropriate. A comparison of continuous variables was performed by Student's t test or Mann–Whitney U test, while more than two groups were compared using either ANOVA or Kruskal–Wallis test. Categorical variables were analyzed using Chi-square with Fisher's exact test. A two-tailed P value of < 0.05 was taken as significant. Statistical Package for Social Sciences (SPSS) (version 23; IBM Corp, Armonk, NY, USA) was used for data analysis.
One hundred and four histopathology-proven craniopharyngioma patients (age: 6 months–17.5 years) were seen in our hospital between 2003 and 2018. Thirty-three patients (32%) were lost to follow-up. Data were available for 62 children, of which 41 (66.1%) were males and 21 (33.9%) were females. Mean age at presentation was 10.4 (4.5) years (males: 11.1 (4.4), females: nine (4.4) years; P = 0.08). Patients were followed up for a median duration of 60 months (34–84 months). Of 62 cases included in our study, the histopathology subtypes were reported in 47 cases, out of which 44 patients (94%) had adamantinomatous and three patients had papillary histopathology.
Clinical characteristics at presentation
The youngest age at presentation was 6 months in our cohort. The median (interquartile range) duration of symptoms at presentation was 6 months (3–12 months). Only one patient was asymptomatic as craniopharyngioma was incidentally detected post brain imaging after a fall from height. Six children were young at presentation (age 6, 15, 24, 36, 48, and 48 months) to effectively communicate.
Among the presenting symptoms, the most common were headache (72.6%), vomiting (59.7%), and visual complaints (33%). Polyuria and polydipsia were present in four patients. Only six patients complained of growth retardation, but on examination, 60% of children had short stature.
The mean height SD score available for 53 children was − 2.3 (1.34). Thirty-two (60.3%) children had a height SD score of less than − 2 SD. Children with delayed/arrested puberty had significantly lesser height SD score compared to pubertal children (−3.25 (1.12) vs. −1.11 (0.73); P < 0.05).
Endocrine deficits were present in 50 (80.6%) patients at the initial presentation. Hypothyroidism and adrenal insufficiency were noted in 27 (43.5%) and 20 (32.3%) patients, respectively, while only four patients had diabetes insipidus (6.5%). Forty-two (67.7%) were prepubertal, five (8%) had normal pubertal development, and 15 (24.1%) had delayed/arrested puberty.
Thirty-nine (62.9%) patients had normal BMI, nine (14.5%) were overweight, and three (4.4%) were obese at presentation. Hypothalamic involvement (based on the radiological criteria) was noted in 59 of 60 patients for which the radiological details were available (Grade 2 in 55 (88.7%) patients). The tumor was confined to the sellar region without any extension in only one (1.6%) patient.
Visual function at initial presentation: Twenty-six (42%) patients had 6/6 vision and 32 (52%) had less than 6/6 vision in either eye at presentation. Eighteen (29%) had normal visual fields and 32 (52%) had field defects at presentation. Details of visual acuity and visual fields were not available in six (10%), and 12 (20%) were not cooperative for testing.
Management details (surgery and radiation therapy)
Hypothalamus-sparing safe gross total resection was the goal wherever possible. Fifty-three (85.5%) children had undergone trans-cranial surgery, and nine (14.5%) had trans-sphenoidal surgery. Fourteen (22.5%) had no residual lesion, and 48 (77.5%) had a residual lesion in the follow-up MRI after the initial surgery. Four patients underwent a second surgery for large residual lesions. As per the management protocol in the institute, all patients (>3 years old) with residual lesions on postoperative MRI are advised to undergo radiotherapy. Radiotherapy was given to 36 patients (75%) with residual lesions. Younger age (combined with small residual lesions) and loss to follow-up to radiotherapy unit post-surgery were the reasons for the remaining 12 patients not undergoing radiotherapy.
Long-term clinical outcomes on follow-up
Endocrine outcomes at follow-up: Patients were followed up for a median of 60 (34–84) months. At least one endocrine deficit was present in all 62 (100%) patients at last follow-up. Hypothyroidism and adrenal insufficiency were noted in 57 (91.1%) and 55 (88.7%), respectively. Nineteen (31%) were prepubertal, only five (8%) had normal pubertal development, and 38 (61%) patients had delayed/arrested puberty. Thirty-one (50%) children had diabetes insipidus. None had recovery of hypothyroidism or adrenal insufficiency. There was no significant difference in the endocrine outcomes between the groups of patients with residual lesions versus patients without residual lesions postoperatively. A comparison of the clinical characteristics at presentation and at last follow-up is presented in [Table 1].
Nine (14.5%) children were diagnosed with growth hormone deficiency (GHD) and were treated with growth hormone for a median duration of 12 months (10–35 months). A significant increase in mean height S.D score was noted after growth hormone therapy (−3.8 (1.4) at start vs. −2.9 (1.2) at last follow up; P = 0.008).
Prevalence of obesity: The prevalence of obesity increased from 3 (4.8%) at presentation to 9 (14.5%) at 1-year post-surgery and to 14 (22.6%) at last follow-up. The prevalence of obesity increased by nearly fivefold at last follow-up (P < 0.01). Moreover, 28 (45%) children had abnormal BMI (overweight and obese) at last follow-up.
Visual function at last follow-up: Twenty-nine (47%) patients had 6/6 vision, and 31 (50%) had less than 6/6 vision in either eye at last follow-up. Nineteen (31%) had normal visual fields, and 37 (60%) had field defects at last follow-up. Details of visual acuity and visual fields were not available/not cooperative for two (3.2%) and six (9.2%) patients at last follow-up, respectively. Ten (16%) children had severe limitations in performing daily activities (age-appropriate) due to visual defects.
There were 13 events of recurrence/progression. Among these, 10 events occurred in patients who had only surgery (n = 26), and three cases were in patients who also underwent radiotherapy. The median time to recurrence/progression was 36 months (7.7–78 months) from the time of surgery.
The rate of tumor recurrence/progression was significantly higher in children who were treated with surgery alone as the initial treatment in comparison to the children who also underwent radiotherapy after surgery (39% vs. 8%; P = 0.002). Postoperatively, the recurrence rate was not significantly different in the patients with residual lesions compared to those without residual lesions.
Of the 104 patients, eight (7.7%) patients died and 33 (31.7%) patients were lost to follow-up. The cause of death could not be ascertained as all eight children succumbed at home.
Pediatric-onset craniopharyngioma (CP) is a rare benign tumor, but it accounts for 5%–10% of childhood intracranial tumors. As most of them are closely related to the pituitary, optic pathways, and hypothalamus, they are invariably associated with varying degrees of pituitary hormone deficiencies, visual deficits, and hypothalamic syndromes. Despite being a benign tumor, recurrences are common even many years after the initial surgery. Data on childhood craniopharyngioma from the Indian subcontinent are lacking. Our study is the first one to present comprehensive long-term outcome data on endocrine deficit, visual deficit, hypothalamic obesity, and recurrences in pediatric-onset craniopharyngioma from the Indian subcontinent.
In our cohort, nonspecific symptoms of raised intracranial pressure, such as headache, nausea, and visual disturbances, were the most common complaints at presentation, which is similar to the finding in previous studies., However, Gautier et al., in a study of 166 children and adults with CP, reported the prevalence of seizures in 15 (25%) of 60 children aged <18 years at presentation, but in our study, only four (6.4%) children had seizures.
Studies report endocrine deficits in 52%–87% childhood CP at presentation, which includes growth hormone (41%–75%), gonadotropins (20%–56%), adrenal insufficiency (8%–68%), hypothyroidism (15%–32%), and diabetes insipidus (17%–27%).,, Prevalence of endocrine deficits was similar in our study except for the very low prevalence of diabetes insipidus, which was seen in only four children (6.5%). The most likely explanation for the low prevalence of diabetes insipidus at presentation may be under-reporting as untreated hypothyroidism and hypocortisolism in the patients at baseline may mask the diabetes insipidus from manifesting at the time of presentation. Interestingly, there were 60.3% children in total with short stature at presentation, but only 9.7% complained of poor height gain. Inadequate awareness in the community regarding optimum height gain combined with poor height recording and monitoring at schools may be the reason for parents being not aware of poor height gain in their respective children.
We found visual deficits in terms of decreased visual acuity and visual field defects in more than half of our patients at the time of diagnosis, which is comparable to the data reported in the literature.
The prevalence of pituitary hormone deficiencies increases during long-term follow-up. Deficiencies of growth hormone (93%–96%), gonadotropins (59%–95%), adrenal insufficiency (78%–100%), hypothyroidism (86%–100%), and diabetes insipidus (65%–85%) have been reported in different studies.,, We also found similar observations of increased prevalence of endocrine deficits post treatment as all 62 children had at least one endocrine deficit at last follow-up. The prevalence of diabetes insipidus in our study was less both at presentation (6.5%) and at last follow-up (50%) compared to the studies mentioned above.
Gross total resection confers a significant risk for the development of new endocrinopathy compared to subtotal resection with/without external radiation. Pituitary hormone deficiencies might also occur later in life, particularly when the patient has received radiotherapy; therefore, monitoring of pituitary function is a life-long requirement.
None of our patients showed recovery of endocrine deficits, though rarely recovery of hormonal deficits after surgery is known to occur. Honegger et al. studied endocrinological outcomes of craniopharyngioma surgery in 143 patients (30 were <16 years old; the rest were adults) followed for a median of 60.2 months and noted that 88 patients remained free of recurrence and did not need radiotherapy. Nine of the 88 (10.2%) patients had recovery of at least one pituitary deficit (diabetes insipidus = 5, adrenal insufficiency = 3, and hypogonadism = 1).
Physiological doses of growth hormone are safe and do not increase the size and recurrences rates of the tumor. GH treatment is essential to treat GHD and short stature. Significant improvement in height SDS was noted in nine children who received GH treatment in our study for a median duration of 12 months (P = 0.008). None of these nine children had an increase in tumor size during GH treatment.
Hyperphagia and decreased resting energy expenditure due to hypothalamic damage as a consequence of the tumor or the surgery cause hypothalamic obesity. Physical inactivity due to visual deficit and other morbidities associated with tumors contribute to obesity. Various studies have reported the prevalence of obesity at presentation around 12%–19%. Following treatment, the prevalence of obesity may increase up to 40%–60%.,, Sterkenburg et al. compared BMI among 77 children with (n = 52) and without (n = 25) hypothalamic involvement at three time points—at diagnosis, during 8–12 years after diagnosis, and after more than 12 years of diagnosis—and noted that BMI increase and consequent development of hypothalamic obesity stabilized, plateauing after 8–12 years of follow-up in both groups.
Despite the majority having hypothalamic involvement by the tumor, the prevalence rates of obesity noted in our study (4.8% and 22.6% at presentation and at last follow-up, respectively) were considerably less than those reported from Western literature.,, As it is observed that BMI increases progressively in the initial 8–12 years after diagnosis, the lower prevalence noted in our study may be due to a shorter period of follow-up time (median of 60 months (34–84 months)).
Visual deficits and impairment of visual acuity were reported to exist in more than 50% of patients at diagnosis, which is similar to the observations in our study. While very few patients had improvement in vision after surgery, it remained stable in most of the patients and even worsened in some due to progression/recurrence or surgical insult. Visual defects led to severe limitations in performing daily activities (age-appropriate) in 10 (16%) children.
Post-surgical improvement in vision may occur in 41%–48% of patients., However, we did not observe visual functions improvements in our patients except for a couple of cases. The risk factors for non-improvement in the visual function post-surgery include severe pre-surgical visual deficits and tumor localization in the pre-chiasmatic area., In our patients, the two most probable factors for non-improvement were severe visual deficits at presentation and delayed surgical decompression due to late medical attention as well as a longer waiting time for surgery sometimes.
Radical resection of the tumor is associated with surgically induced hypothalamic and visual deficits. There is an increasing trend toward less radical, hypothalamic sparing surgeries in recent years.
As per the experience-based management protocol in our institute, safe gross total resection with hypothalamus-sparing approach should be the goal wherever possible. For tumors confined to sella (grade 0), the chances of gross total resection are high. The surgical corridor/extent of surgery is influenced by many clinical as well as radiological factors such as radiological tumor extent, hypothalamic involvement/adhesions, pneumatization of the sphenoid sinus (for transsphenoidal surgery), urgency of surgery (decompression vs. excision) in cases of sudden rapid vision loss or neurological deterioration, presence and severity of hydrocephalus, and preoperative panhypopituitarism.
Incomplete resection without radiotherapy is associated with high rates of residual tumor progression, whereas the rate of progression with radiotherapy is much less. We also had a similar finding of significantly less recurrence or progression in children who received radiotherapy in comparison to the children who did not. In one child, the longest time period noted for recurrence was 108 months, which means careful surveillance is needed for a much longer period of life for early detection of recurrence/progression. Müller et al. reported in the results of the multicenter prospective trial, KRANIOPHARYNGEOM 2000, that irradiated patients had an 88% lower risk of recurrence/progression compared to patients without/before irradiation (hazard ratio = 0.12; P < 0.001). However, the optimal time for radiation therapy after incomplete resection is still not clear.
The current study is one of the largest series of childhood-onset craniopharyngioma from developing countries. An attempt was made to regularly follow all the patients. However, our study had a few limitations. Because it was mainly a retrospective study, all investigations and anthropometry were not available for every patient at every visit. Another limitation is the shorter median follow-up time of 60 months. In addition, despite an attempt to ensure follow-up by either clinic visits or telephonic conversations, some patients were not traceable.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.