Comparison of Physiologic Growth Hormone Replacement (n=14 ...€¦ · Begg’s rank correlation...

Results

Objectives

Methods

A systematic review and meta-

analysis were conducted to

examine the effect of GHRT on the

recurrence of craniopharyngioma

among pediatric patients who

underwent a surgical resection.

Conclusions

Discussion

Strengths and Limitations

Comparison of Physiologic Growth Hormone Replacement

Therapy with No Replacement on the Recurrence of

Craniopharyngioma in Pediatric Patients: a Meta-analysis.

Nawaf M. Alotaibi†, B.Pharm, MSc2, Nadia Noormohamed†, Pharm D2, Salman Alharthy, MSc, B.Pharm2, Joanne Doucette, MS2, Hassan Zaidi, MD1, Rania A. Mekary‡, MSc, PhD1,2, Timothy R. Smith‡, MD, PhD, MPH1,2

†Co-first authors; ‡Co-senior authors1Brigham and Women’s Hospital, Boston, Massachusetts; 2MCPHS University, Boston, Massachusetts

Introduction

•Surgical resection of

craniopharyngiomas can result in

hypopituitarism

•In the pediatric population,

postoperative growth hormone

deficiency can result in an

impaired organ maturation.

•Significant controversy exists on

whether growth hormone

replacement therapy (GHRT)

contributes to the recurrence of

craniopharyngiomas in the

pediatric population and may

place patients at a higher risk of

residual tumor growth.

•PubMed, EMBASE, and Cochrane

databases were searched through

October 28, 2015 for comparative

studies (n≥ 5 patients) that

evaluated the effect of GHRT on

the recurrence of pediatric

(newborn- 18 years old)

craniopharyngiomas. Pooled effect

estimates were calculated using

fixed- and random-effects models.

•Heterogeneity, in the analysis,

was evaluated by the Q statistic

and I2 statistic.

•Potential sources of heterogeneity

were explored using sub-group

analyses by categorical covariates

(growth hormone replacement

therapy, country, international

journal Impact factor, Newcastle-

Ottawa Scale –to assess study

quality-; and center).

•Covariates such as age range and

type of surgery could not be

assessed for heterogeneity

sources as information was NA.

•Potential publication bias was

assessed by using funnel plots,

Egger’s linear regression test, and

Begg’s correlation test.

•If publication bias was indicated,

the # of missing studies in a meta-

analysis was evaluated by the trim

and fill method.

Articles were found in PubMed (n=404)

Articles were found in Embase (n=694)

Articles were found in Cochrane Library (n=2)

Total Articles (n=1100)

Ide

nti

fica

tio

n

Duplicates articles (n=303)

Articles screened (n=797)

Scre

en

ing

(n=719) articles were excluded as the title or abstract was not relevant

Full-text articles assessed for eligibility (n=78)

Elig

ibili

ty

In

clu

de

d

(n=56) Not English (n=5)

Conference Abstracts and review articles.

(n=10)

Age > 18 years (n=11)

No information about recurrence and

GHRT (n=22)

Not craniopharingoma (n=3)

Case report (n=2)

Errata (n=1)

Comment to article (n=1)

Inter Library Loan (ILL) (n=1)

Articles included in qualitative synthesis (n=22)

Articles included in quantitative synthesis (meta‐analysis) (n=8)

(n=14) Abstracts and missing data (n=8)

Can’t extract event number (n=2)

Mix adults with children (n=1)

Erratum to another study (n=1)

Not for Craniopharingoma (n=2)

Figure 1: Study selection process of the identified articles

Author, Year

Journal Impact Factor Study Design Country Center

Intervention arm n/N for

GHRT

Control arm n/N for no

GHRTAge range,

year % Male Surgery typeStudy duration (Year X - Year Y)

Intervention type (Name[HGH vs rGH], Dose & Frequency) Frequency

Route of Admin GH

Time after diagnosis when GH treatment

was started

GH treatment duration

Newcastle-Ottawa

Scale

Alev Ozon, Z, 2015 1.538Retrospective

case seriesOutside the USA

Single 11/24 na na naGross totalresection

na GH na na 2.7 years na 2

Arslanian, S, 1985 7.148Retrospective

cohortUSA Single 4/13 0/5 3.9-17.8 na

Subfrontal and transphenoidal

1970-1983Somatotropin,

0.1 unit/kg.3x/week IM/SC na na 6

Darendeliler, F, 2006 1.674Retrospective

case seriesUSA Multiple 121/1038 na 4.3-14.9 N/A Not specified

Start on Jan 2004

Genotropin , 0.17 mg/kg/wk

NA IM/SC na 1.9 yr 3

Kanev,P. M, 1991 3.737Retrospective

case seriesUSA Single 0/5 na na na Subfrontal 1985 - 1987

Protropin,0.1 mg/kg each, or daily injections of 0.05 mg/kg.

3x/week SC na na 3

Moshang T, 1995 9.392Retrospective

case seriesUSA Single 2/19 na 3.5-12.5 84.2 Not specified 1985-1995 rHGH na na 2.4 years na 3

Moshang T, 1996 3.79Prospective case series

USA Multiple 35/546 na na na Not specified na rHGH na na Not specified na 3

Price DA, 1997 2.48Retrospective

case seriesOutside the USA

Multiple 54/488 na 4.9-14.8 57 Not specified 1988-1996rGH, mean dose 0.49+-

0.15 IU/kg/6x/ week

(range 3-7)Injection

Median 1.56 yrs , mean 2.23+-1.88 yrs

Some patients after one year

and other patients after 5 years on GH.

2

Rohrer, RT, 2010 0.995Retrospective

case seriesOutsidethe USA

Single 11/22 4/7 1.3-15.5 65.5 Not specified na hGH, 0.025 mg/kg Daily Injection Not mentioned na 2

Table 1: The characteristics of the studies included in the meta-analysis.

•Eight studies (7 case series and 1 retrospective cohort) with 2,167

patients with craniopharyngioma were included in the final analysis.

•The fixed pooled prevalence was 11.7% (95% CI: 10.4%; 13.2%) for

GHRT+ and 43.1% (95% CI: 16.7%; 74.1%) for GHRT- (P-

interaction = 0.23)

• Among the GHRT group, the fixed pooled prevalence was 15.5%

(95% CI: 12.4%; 19.1%) for studies outside the USA and 10.4%(

95% CI: 9.0%; 12.0%) for studies conducted in the USA (P-

interaction= 0.07).

•The fixed pooled prevalence was 10.4% (95% CI: 9.10%; 11.8%) for

multiple-center studies and 37.9% (95% CI: 27.3%; 49.7%) for

single-center studies (P-interaction<0.01).

•The fixed pooled prevalence was 7.50% (95% CI: 5.60%; 10.0%) for

high impact factor journals and12.9% (95% CI: 11.3%; 14.7%) for

low impact factor (P-interaction=0.16).

•The fixed pooled prevalence was 10.4% (95% CI: 9.0%; 12.0%) for

high NOS and 15.5% (95% CI: 12.4%; 19.1%) for low NOS (P-

interaction=0.07)

Figure 2: Craniopharyngioma recurrence prevalence for growth hormone replacement

therapy (GHRT) and no GHRT.

Forest plot represents the craniopharyngioma recurrence prevalence (95% CI) after GHRT

(GHRT+) or no GHRT (GHRT-) among patients with craniopharyngioma who underwent a

surgical resection. Horizontal lines denote 95% CIs; solid squares represent the point

estimate of each study and the diamond represents the pooled estimate of the intervention

effect. The size of the solid squares is proportional to the weight of the study. (I2 for GHRT

+=88.6%; p-heterogeneity<0.01; I2 for GHRT- = 61.7%; p-heterogeneity=0.11 ; overall I2 =

87.3%; p-heterogeneity<0.01 (P-interaction comparing the two groups=0.23).

The fixed pooled prevalence is 11.7% (95% CI: 10.4%; 13.2%) for GHRT+ and 43.1% (95%

CI: 16.7%; 74.1%) for GHRT-. The overall prevalence is 11.9% (95% CI: 10.5%; 13.4%).

Figure 3: Funnel Plot of Standard Error by Logit event rate.

The vertical solid line is drawn at the pooled log odds ratio, and the other two

lines represent the expected 95% confidence interval for a given standard

error. Visually, the Funnel plot showed a slight asymmetry. Statistically, both

Begg’s rank correlation test (P = 1.00) and Egger’s linear regression test (P =

0.24) indicated no significant publication bias. Conducting the trim and fill

method resulted in an overall imputed effect estimate that dropped from 11.9

% (10.5%; 13.4%) to 11.1% (9.87%; 12.5%).

• GHRT among pediatric patients with biochemical

evidence of GH deficiency after surgical resection for

craniopharyngiomas remains a topic of controversy for

clinicians and surgeons.

• Our current meta-analysis demonstrated that pediatric

patients treated with GHRT after surgical resection of

craniopharyngioma had a lower incidence of tumor

recurrence.

• There was insufficient evidence in the current

literature to indicate that GHRT promoted growth of

craniopharyngiomas after surgical resection.

Pros Cons

Explored different sources of

heterogeneity for which the

information was available

Only studies in English were included

Publication bias was assessed Age and perioperative medical treatment modalities

were not available

Heterogeneity remained relatively high in the

subgroups reflecting some residual effect modifiers

that were not reported in all the studies.

• This meta-analysis showed that GHRT use

was not associated with the recurrence of

craniopharyngioma among carefully selected

pediatric patients

Contact Information

Email: rania.mekary@mcphs.edu