The use of PPIs and H2RAs among children has substantially increased in recent years (Hales et al., 2018). PPIs and H2RAs have shown to be effective drugs in children as well as in adults with acid-related gastrointestinal disorders, such as Helicobacter pylori infection and gastroesophageal reflux disease (GERD) (Herszényi et al., 2020; Perry et al., 2020). These drugs are generally considered safe, but children are more vulnerable to drug toxicity (Ward and Kearns, 2013). Accordingly, it is critical to clarify the safety of PPIs or H2RAs in children.

In adult studies, PPIs use can significantly increase the risk of fracture, particularly fractures of the spine and hip (Eom et al., 2011a; Kwok et al., 2011; Ngamruengphong et al., 2011). Recent studies have suggested that PPIs may also increase the risk of fracture in children and young adults; there are several biological mechanisms that explain a possible relationship between the PPIs use and the risk of fracture. Some studies found that PPIs can reduce calcium absorption through the inhibition of gastric acid secretion, which leads to increase of the risk of bone fracture (Yang et al., 2006; O’Connell et al., 2005). PPIs may decrease the absorption of vitamin B12 (Lewis et al., 2014), which is associated with bone mineral content and density in elderly women (Dhonukshe-Rutten et al., 2003). Moreover, PPIs might have direct deleterious effects on osteoclast and osteoblast cells, which might result in a reduction in bone turnover (Costa-Rodrigues et al., 2013).

H2RAs are an alternative medication to treat GERD in children and young adults (Cohen et al., 2015). One study showed that H2RAs are safer for the treatment of GERD in children and adolescents (Ruigómez et al., 2017), and there was no significant risk of fracture with H2RAs use (Kwok et al., 2011). However, another study reported a high risk of fracture among young adults receiving H2RAs (Freedberg et al., 2015a). Whether the use of H2RAs or PPIs can significantly increase the risk of fracture is still much debated. A number of studies have investigated the association between the PPIs or H2RAs use and the risk of fracture in children and young adults, but showed inconsistent results. To our knowledge, there is no comprehensive meta-analysis of the related studies yet. Therefore, we conducted such meta-analysis to identify the association between PPIs and/or H2RAs use and the risk of fracture in children and young adults.

The MOOSE guidelines were adhered for meta-analysis of observational studies and PRISMA guidelines (Stroup et al., 2000).

We initially identified 114 potentially eligible studies after the literature search process, and after removing 31 duplicated ones, 83 studies remained, in which we excluded 57 other studies which were obviously irrelevant based on the titles and abstracts. Also, after full-text assessment, we eventually identified six studies that could be used for our meta-analysis. The literature search process is illustrated in Figure 1.

In this meta-analysis concerning the six selected observational studies, we found an increased risk of fracture with the PPIs use alone or the PPIs combined with H2RAs use in children and young adults, but not with the H2RAs use alone, which is consistent with the meta-analysis results of the acid-suppressing medication and fracture risk in adults (Kwok et al., 2011; Freedberg et al., 2015a). Besides, the use of PPIs or H2RAs increased the risk of fracture more significantly in one case-control study as compared to the cohort studies. However, it is generally considered that cohort studies provide superior evidence compared with case-control studies because cohort studies are less susceptible to bias in the prospective design. As a result, the results obtained from this meta-analysis should be interpreted with caution. Interestingly, the subgroup analyses by sex showed that the PPIs use increased the risk of fracture in women, but not in men. The subgroup analyses by fracture outcome showed an increased risk of spine, low limb, and other fracture for the PPIs use. The region of the study conducted may be one of the sources of heterogeneity among the studies. After limiting to North America, there was low heterogeneity across the related studies (I ² = 0.0%, p > 0.05).

PPIs use is becoming increasingly common, and there has been an observed increase in the risk of gastrointestinal and respiratory tract infection (Eom et al., 2011b; Freedberg et al., 2015b), hypomagnesaemia, vitamin B12 deficiency, and bone fracture (De Bruyne and Ito, 2018). Particularly, the use of PPIs or H2RAs might increase the risk of bone fracture and osteoporosis (Targownik et al., 2010; Liu et al., 2019). The risk of fracture with PPIs use was higher than that with H2RAs use, which might be explained by a few reasons. First, PPIs have a stronger inhibitory effect on the acid production than H2RAs; the latter only suppress 70% of gastric acid production (Colin-Jones, 1995), whereas the former, PPIs, block 98% of the acid production (Schuler, 2007). Insoluble calcium salt from diet is difficult to absorb (Insogna, 2009), and the ionized calcium from insoluble calcium salts is promoted to release in the acidic environment of the stomach; thus, the reduced acid production could impair calcium absorption (O’Connell et al., 2005). Second, it was also reported that calcium malabsorption would cause hyperparathyroidism, thus leading to the reduction of bone mineral density (Insogna, 2009). Third, PPIs use could reduce vitamin B12 absorption, which may increase the risk of fracture as well (Marcuard et al., 1994). Fourth, cimetidine, one of the H2RAs, has been shown to increase bone mineral density by preventing osteoclast differentiation induced by histamine (Yamaura et al., 2003). Similarly, Vesterdaard et al. (2006) found that PPIs use might increase the risk of fracture whereas the H2RAs use decreases the risk. Although PPIs use was associated with increased risk of fracture, some studies found that there was no significant difference between the PPIs use and the level of bone mineral density (Targownik et al., 2010; Targownik et al., 2012; Hansen et al., 2019). Last, PPIs might increase the incidence of nonosteoporotic fracture by increasing the risk of falling (Thaler et al., 2016).

Our meta-analysis has several strengths. First, there were several previous meta-analyses on the fracture risk with PPIs or H2RAs use in the elderly. However, the current meta-analysis is the first to review the PPIs use and fracture risk in children and young adults. Second, it examined the associations stratified by the type of agent (PPIs or H2RAs), the study design, region, NOS score, outcome, and sample size, as well as sex. However, our meta-analysis also has several limitations. First, there was no available RCT on the issue, so our study consisted entirely of observational studies, which are also susceptible to various biases. Second, the number of eligible studies selected in our meta-analysis was relatively small. Third, some potential studies might have been missed with the language of publications restricted to English only.

In summary, this meta-analysis of observational studies showed that PPIs use alone could increase the risk of fracture in children and young adults, but not for PPIs combined with H2RAs use or H2RAs use alone. Clinicians should carefully assess the risk of fracture before prescribing PPIs for patients. Further studies are required to confirm our findings reported hereby due to the inevitable limitations of this meta-analysis.