This study analyzed outcomes of patients with MPS who underwent AT. Patients with MPS tended to be younger, have a longer length of stay, and have a higher total cost of hospital stay than patients without MPS.

We also found a significantly increased length of hospitalization and cost of care in patients with MPS, which is congruent with findings from previous studies for other procedures. More specifically, a total cost-analysis study by Davari et al. [29] recorded that the average annual cost of treatment for MPS-I patients is $87,971.99, a quantity 16.2 times the GDP per capita in Iran. A separate cost-analysis study by Conner et al. [30] recorded an average cost per patient per month of $10,715 for an average annual cost of $128,580 for medical care of patients with severe MPS-I. It is important to note that neither of these studies specifically investigated surgery-specific costs of care. The study by Davari et al. includes surgical costs in its estimate of the annual cost of care for MPS-I patients but does not provide a specific annual estimate of surgical care per patient. The study by Conner et al. only recorded costs for medical interventions and indicated surgical costs as an area for future study. Given the paucity of information in the available literature, quantifying the cost of surgical management of MPS remains a viable area for future study. However, when comparing MPS to other complex medical conditions, the general trend in the literature, which consistently demonstrates higher costs of care for children with complex medical conditions and congenital syndromes, applies directly to children with MPS both in terms of medical management as evidenced by the above studies, and from a surgical perspective as reported in our retrospective review of the HCUP KID database for patients with MPS undergoing AT.

Interestingly, there was a statistically significant trend toward an increased risk of post-tonsillectomy hemorrhage (PTH) in patients with MPS, with an adjusted odds ratio of 5.97. Increased post-tonsillectomy hemorrhage rates have been documented in other complex pediatric patient populations, such as patients with Down syndrome and cerebral palsy, but this has not been previously reported in MPS. Data on postoperative bleeding in patients with MPS across surgical disciplines is scarce. A PubMed review yielded a lack of studies that have specifically documented the incidence of postoperative hemorrhage in patients with MPS or the association between specific surgical procedures and postoperative hemorrhage in patients with MPS. Furthermore, in patients with certain subtypes of MPS, including MPS-IH (Hurler Syndrome), studies have shown coagulopathy directly related to glycosaminoglycan levels in the blood [31]. Whereas postoperative blood loss in patients with MPS undergoing AT has not previously been studied, this finding was statistically significant in our study. Thus, it is worth being aware of this potentially increased risk in this vulnerable patient population, and providers performing AT on patients with MPS should do so at tertiary care pediatric centers best equipped to manage these potential complications.

There was a significantly higher rate of respiratory complications in children with MPS compared to children without MPS, with an adjusted odds ratio of 6.88. To our knowledge, this finding has not been previously investigated. However, similar postoperative respiratory complications have been found in other children with congenitally complex airways. For example, a recent systematic review by End et al. investigated the risks and benefits of AT in children with cerebral palsy (CP) and OSA. They reported children with CP have a significant risk of developing postoperative respiratory complications, including pneumonia, and that these patients required additional airway management with increased requirement of unplanned ICU admission [32]. This pattern is consistent with our findings, which demonstrate a higher risk of postoperative respiratory complications in children with MPS compared to those without MPS.

To our knowledge, this is the first large database study examining post-tonsillectomy complications in this patient population. However, providers should note several limitations to the current study. The first limitation is the study’s retrospective database design. Further, our comparison group did not only consist of otherwise healthy patients but included all other pediatric patients undergoing AT. This design may have overestimated complication rates, length of stay, and costs in this group. However, our data still achieved statistical significance in several outcomes. MPS contains seven different unique subtypes. The HCUP KID database is based on ICD-10-CM/PCS data only. Therefore, there is a lack of granularity inherent to this database that, unfortunately, did not allow us to account for specific MPS subtypes or disease severity by creating a severity of illness subclass within those particular subtypes.

Furthermore, we could not differentiate between specific subtypes of respiratory complications or between primary versus secondary PTH. Since tonsillectomy status and the procedure code for PTH were part of the admission criteria, providers may assume that all instances of hemorrhage in post-tonsillectomy patients included in the HCUP KID database are primary hemorrhages. Similarly, polysomnographic data or information on OSA severity was not available. Patients with more severe OSA are at higher risk for postoperative complications, and due to the lack of this data, we were unfortunately not able to control for this confounder. There may be a difference in OSA severity between patients with MPS and other patients studied herein. However, given the nature of the HCUP KID database, we do not know this to be true without a doubt. However, the data collected from the HCUP KID database is all inpatient data. Therefore, all patients included and compared in the study were hospital inpatients. Apart from age, the most common reason for postoperative admission would be due to severe OSA or other comorbidities, such as congenital syndromes that predispose patients to severe OSA.

Furthermore, among children with severe OSA, there is no substantial evidence to support that increasing AHI is correlated with an increased risk of PTH. Therefore, the striking difference between patients with MPS and patients without MPS in our study is unlikely to be solely due to differences in OSA severity. In addition to disease severity, the HCUP KID database did not allow for granular analysis of additional potential confounders known to be associated with a higher risk of PTH, including but not limited to infection, hematologic diseases, and neurodevelopmental diagnoses. While our patients did not have additional hematologic diseases that are not inherent to MPS, the effects of MPS on the hematologic system are well-known and established. The inherent hematologic dyscrasias of MPS may be a confounder that was not adjusted for when considering PTH in patients with MPS. However, this increased risk is inherent to all patients with MPS.

Additionally, patients with active infections or poor functional status, on average, have been found to have increased rates of post-tonsillectomy complications compared to other patient populations, such as patients with cerebral palsy [33]. We did not have this data available through the HCUP KID database to address all confounders fully. Lastly, the proportion of male-to-female children was higher in the MPS group. There is evidence of an increased risk of PTH in male patients in the literature, and this could have been a confounding factor. However, our odds ratio was adjusted for sex and continued to demonstrate the robustness of the results of the current study.