Linear growth after enterocystoplasty in children and adolescents: a review
- First Online:
- Cite this article as:
- Mingin, G., Maroni, P., Gerharz, E.W. et al. World J Urol (2004) 22: 196. doi:10.1007/s00345-004-0433-9
- 75 Downloads
The interposition of bowel in continuity with the urinary tract has allowed for the preservation of renal function and continence in children with bladder exstrophy, as well as neurogenic and valve bladders. Although bladder augmentation with ileum or colon has been shown to be safe, the long-term effects of metabolic acidosis in addition to abnormalities in linear growth and bone metabolism remain largely unknown. We reviewed the literature to critically examine linear growth in children who have had bladder augmentation with a particular emphasis on the correlation between acid-base status, bone mineralization and growth. The majority of studies suggest that linear growth is not affected by bladder augmentation. In the short-term, children post-augmentation have varying degrees of metabolic acidosis which, overtime, appears to resolve with no affect on linear growth. In a single study, bladder augmentation led to significant bone demineralization almost a decade after surgery, however, even in these children no decrease in linear growth was noted. No alterations in bone density levels were seen with short-term follow-up.
KeywordsLinear growth Enterocystoplasty Urinary diversion
The interposition of bowel in continuity with the urinary tract was first described more than 150 years ago . Coincident with this was the first description of rickets in a patient after urinary diversion . With the discovery by Lapides that intermittent catheterization allows for a safe and effective method to empty the bladder, intestinal augmentation procedures and continent intestinal reservoirs have become the standard of care for pediatric urology patients . The use of bowel within the urinary tract has allowed the preservation of renal function and provided urinary continence in children with bladder exstrophy, neurogenic, and valve bladders. Intestinal interposition has not been without complications. Metabolic derangements were apparent after the initial interposition of bowel and have been widely reported since Boyd’s observation of osteomalacia in 1931 . Although extended follow-up of children with bladder augmentation has been shown to be safe, the long-term effects of chronic metabolic acidosis remain unknown . Abnormalities in growth and bone metabolism have also been a concern in patient’s with intestinal interposition in their urinary tract. Over the last decades there have been several reports of delayed linear growth in children whom have undergone bladder augmentation with ileum or colon [2, 4, 9, 14]. More recently, decreasing bone density has been reported in a similar patient population . The current literature is contradictory in assessing the extent of metabolic derangement, changes in linear growth and bone mineral density. The goal of this review is to critically examine growth in children who have had a bladder augmentation. Although focusing on linear growth, this topic can not be viewed in isolation, separated from acid-base status and changes in bone mineralization.
Study populations and methods
The recent literature that has investigated the potential adverse effects of bladder augmentation (with the exception of Mingin et al.  and Kockum et al. ) focuses on children with bladder exstrophy and neurogenic bladder to the exclusion of myelomeningocele. This may in part be due to the knowledge that normal growth is impaired in patients with spinal defects. However, the assumption that bladder exstrophy patients have normal growth has now been challenged. Sponsellar et al. reported their studies of bony abnormalities in children with the exstrophy complex ; of note they describe a 30% shortage of bone as well as undergrowth of the pelvis. The authors believe that undergrowth of the pelvis and external rotation of the pelvis and hips may cause a linear increase in joint load and stress affecting the growth plates of the long bones of the lower extremities. Moreover, Feng et al.  found the exstrophy patients in their control population to be of shorter stature then their augmented patients. Thus, as Gearhart has stated, if the control group has a tendency toward shorter linear stature any effect of augmentation will be additive . The fact that both myelomeningocele and now exstrophy patients appear to have growth abnormalities of the lower extremities raises questions concerning our ability to accurately measure linear growth. In most studies, heights were determined through a retrospective review of growth charts. Inconsistencies arising from this include a paucity of both pre-operative and post-operative measurements as well as a failure to describe how the measurements were obtained. Percentile height calculations, dependent on height data points, measured from head to toe, no matter what statistical package is used to control for spurious measurements, will be inaccurate. With the understanding that both of these groups of patients cannot be compared to normal children, comparisons among augmented myelomeningocele and exstrophy patients to non-augmented myelomeningocele and exstrophy patients appear valid.
Linear growth results from a complex process of integrated physiological mechanisms and may be disturbed by a number of these factors acting either independently or together. Although there is no doubt that failure of normal growth can result from a wide variety of organic diseases, it may be extremely difficult to establish such a causal connection, particularly in a retrospective study. Apart from the well described pitfalls and imprecisions in measuring height and calculating growth velocity there is no exact definition of failure of normal growth in an individual because such a definition varies according to clinical and epidemiological needs. There seems to be a consensus, however, that the shorter a child’s stature, the more likely it is that he/she is failing to grow normally. While growth velocity is liable to misinterpretation, a diagnosis of abnormal growth requires long-term monitoring and is best seen as series of height measurements crossing the centiles on the height chart.
Mundy and Nurse and Wagstaff et al. [9, 14] were the first to study linear growth in children post bladder augmentation. In the first of these studies, heights were obtained in 16 children. Three of these children post colocystoplasty were reported to have a 20% reduction in growth. The remaining 13 children, ten of whom had an ileocystoplasty, showed no change in growth. In the second study Wagstaff et al. measured heights in 60 patients. Twelve (20%) patients were reported as having a delay in linear growth defined by a change in percentile height post intestine-cystoplasty. In subsequent follow-up, Wagstaff continued to report a 20% delay in linear growth with accelerated growth in nine (15%) of patients.
Ten years after the original study, Gerharz et al. re-evaluated the 12 children who had impaired linear growth only to find that when compared to their percentile position in the original study two showed no change while ten improved their position . Of these, four had surpassed and two had regained their preoperative percentile.
More recently, only two studies report a decrease in linear growth, while five report no change. Gros et al.  reported delayed growth as defined by a postoperative decrease in percentile height in 14 of 17 (82%) augmented exstrophy patients. This represented a decrease of 15.6 percentile points between pre and post-augmentation. Notable in this study was that delayed growth occurred in 33% of the non-augmented exstrophy controls.
In addition, Feng et al.  showed a similar loss of 15.3 percentile points in augmented children with classic bladder exstrophy. Interestingly, this group also reported delayed linear growth in their control population.
The majority of studies appear to contradict the above findings. In one of only two studies to measure linear growth in both the myelomeningocele and exstrophy population, Mingin et al.  reported no change in linear growth among 22 augmented myelomeningocele patients and 11 augmented exstrophy children when compared to controls. Similarly, in a study of 15 children, five with bladder exstrophy and ten with neurogenic bladder (nine having myelomeningocele and one with sacral agenesis), there was no growth impairment reported . In a study excluding myelomingocele patients, Gerharz et al.  reported that after enterocystoplasty 85% of their 123 patients remained the same or reached a higher percentile. Of the 15% who showed a decrease, 93% remained above the ninth percentile and in 3.3% a clinically relevant growth disorder was recognized. A further study by the Toronto group , again looking at 25 patients with a diagnosis of bladder exstrophy, determined that none of the augmented children had an alteration in supine height measurement. Finally, in a study of 28 children with a mixed diagnosis (myelomingocele is not specifically mentioned) who underwent enterocystoplasty, no change in growth was reported in children augmented with colon. In those who received a sigmoid cystoplasty the average linear growth decreased during the first year postoperatively, but recovered to preoperative percentiles 2 years after surgery .
Among those studies which showed a decrease in linear growth, serum bicarbonate levels were found to differ by only 1.6 mmol/l in children having decreased growth (mean follow-up 9 years) , while Gros et al.  reported no significant metabolic disturbances among augmented children (mean follow-up 5.7 years).
In the studies that did not show a change in growth, two reported no metabolic disturbances with a follow-up of 3.7 and 10 years, respectively [5, 8]. In contrast, significant differences in serum bicarbonate and chloride were seen in children augmented with ileum colon and sigmoid as reported by Mingin et al. and Vajda et al. These studies had a follow-up of 3.7 and 6 years, respectively [8, 13].
Bone mineral density
Only three recent studies have assessed bone mineral density in children post surgery [5, 6, 8]. In each of these studies, no change in linear growth was found post augmentation. In all three of the studies, dual photon x-ray absorption (DEXA) scanning was used to obtain bone mineral density measurements. Both Mingin et al.  and Kockum et al.  reported no significant change in bone mineral densities in augmented myelomeningocele/exstrophy patients. The average length of follow-up in both these studies is 3.7 years. In contrast, Hafez et al.  report a mild reduction in bone density between 1 and 2 standard deviations in 12% of their augmented exstrophy patients, with severe osteopenia (2 or more standard deviations) seen in 20% of patients. The average length of follow-up in this study is 8.9 years.
We have presented an overview of the most recent literature describing the effects of intestinal cystoplasty on linear growth, acid-base status and bone demineralization. Through careful examination several points become clear.
First, does augmentation affect linear growth? If we examine the two earliest studies only Mundy et al.  showed a reduction, and this occurred in just three patients. Given that this was a retrospective study with an unclear definition of growth rate, the results remain provocative yet unsubstantiated. Although Wagstaff  initially reported growth failure in 12 patients, subsequent follow-up 10 years later refuted the initial finding . When the seven most recent studies are reviewed, five show no change in linear growth. Although two out of the five have a relatively short follow-up (3.7 years) the other three have an average follow-up ranging from 6 to 16 years. In the studies reporting a decrease in linear growth, there are two potential problems. Combining both of the studies, only 35 patients were evaluated compared to a total of 227 in the five studies showing no change in growth. In addition, the number of height measurements is not recorded in one  while the other mentions that at least one preoperative and postoperative height was obtained. In contrast, Mingin et al. and Gerharz et al. recorded a minimum of eight heights in the former and ten in the latter. Although not completely conclusive the literature would suggest that, to date, growth may not be affected to the extent reported by Feng et al. and Gros et al.
The questions: “does augmentation lead to a change in acid-base status?” and “does this impact on linear growth?”, can be answered in two ways. In the short term, it appears that children augmented with intestine have varying states of metabolic acidosis as seen through low serum bicarbonate levels . In the long-term the acidosis appears to be in a compensated state [2, 5], quite possibly due to respiratory factors in combination with bone buffering. Either way, the initial acidosis does not appear to affect long term growth.
Finally does bladder augmentation affect bone mineral density and does this lead to a subsequent growth defects? Although none of the studies in which bone density was obtained show a decrease in linear growth, they do suggest a possible disturbing trend. Two of the studies show no change in bone density post augmentation [6, 8], however, the follow-up for each was only 3.7 years. In the study with 32% of their patients having significant bone demineralization the follow-up was almost 10 years . This interpretation must be viewed with caution since this is a single study based on a relatively small number of patients.
Overall, it does appear that intestinocystoplasty leads initially to a state of significant metabolic acidosis, as time progresses the body compensates with bone buffering, at least in part. This explains why over the long-term acidosis resolves at the expense of decreased bone mineralization. Either initial acidosis or bone demineralization may have some effect in stunting short-term growth as reported by Wagstaff et al. and Vajda et al. Interestingly, decreased bone mineralization does not appear to affect final adult height. Why this is so remains a mystery. Acidosis appears to set in motion the breakdown of bone. Overtime demineralization allows for compensation of the acidosis, however, it also leads to an increase in both osteoclast and osteoblast metabolic activity. Increased osteoblast activity is reflected in an increase in osteoblast-specific enzyme activity that has been reported in children post augmentation . It is possible that bone demineralization is the price paid for metabolic compensation and continued normal or near normal linear growth.
The last questions remaining are the long-term significance of bone demineralization, and can these effects be altered by early therapeutic intervention in the presence of normal bone densities. Currently there are no studies to suggest or refute the significance of how osteopenia will affect these children as adults. Clinically, we do not see pathological fractures in our young adult patients; however, only time will tell. In those patients who are non-ambulatory, the effects may be less severe than in patients who are weight bearing.
In summary, these conclusions are based on a small number of studies. From the available data, it appears that bladder augmentation does not affect final adult height. In the initial years after augmentation, most patients experience a metabolic acidosis which is corrected over time. Although based on a single study with 10 year follow-up, it is reported that over the long-term bone densities decrease. Until there is an acceptable replacement for bowel interposition, long-term prospective studies are needed to definitively prove that bladder augmentation has a detrimental effect on bone mineralization.
The study by Mingin et al. 2002  was supported by NIH M01 RR01271 from the UCSF Pediatric Clinical Research Center.