**3. Materials and Methods**

A systematic review of the literature was performed through PubMed, ISI Web of science, and Cochrane Library. "Scurvy", "vitamin C deficiency", "Moeller's disease", "Cheadle's disease", "scorbutus", "Barlow's disease", "hypoascorbemia", "lack of vitamin C", "scorbutic", "child", "children", "pediatric", "toddler", "infant", "infancy" and "childhood" were employed as Medical Subject Headings (MeSH) terms (Table 1). Search operations have been completed in December 2020. The PICOS criteria and PRISMA checklist have been fulfilled in the review execution [18,19]. Male and female children (age ≤ 16 y), with confirmed diagnosis of scurvy, characterized by any kind of clinical manifestation related to vitamin C deficiency was set as population of interest (intervention); "no intervention" was comparison. Comparative studies, cross-sectional studies, retrospective studies, prospective studies, survey studies, case series and case reports were included. We aimed to identify the clinical manifestations and diagnostic methods of scurvy. The inclusion criteria of the selected studies were: presence of any clinical manifestation in humans, pediatric age, edited in English language, published since 1990. Review articles and studies without full text available were excluded. Three reviewers, once the initial results were collected, analyzed the titles and abstracts; then duplicates were excluded, and all those articles that did not match the inclusion criteria were ruled out. The full texts of the remaining articles were read in depth by two reviewers to better assess the content of the studies: demographic data, clinical manifestations, diagnostic path, therapeutic approach and all the extracted data have been organized in Table S1. The Risk of Bias in Non-randomized Studies of Interventions (ROBINS I) assessment tool was employed to evaluate quality of non-randomized studies [20]. This tool analyzes seven bias domains and each one refers the Risk of Bias (RoB) in five grades: low (LR), moderate (MR), serious (SR), critical (CR) and no information. The overall evaluation is based on the combination of these seven domains [21] (Table S2). A study based on a non-randomized design rarely presents a low level of RoB. The review was submitted and registered on PROSPERO [22] (registration code: CRD42021225174).


**Table 1.** The entire list used in the search and the combinations used in the research phase.

1. Scurvy; 2. Vitamin C deficiency; 3. Moeller's disease; 4. Cheadle's disease; 5. Scorbutus; 6. Barlow's disease; 7. Hypoascorbemia; 8. Lack of vitamin C; 9. Scorbutic; 10. Child; 11. Children; 12. Childhood; 13. Infant; 14. Infancy; 15. Toddler; 16. Pediatric.

#### **4. Results**

## *Systematic Review of the Literature*

The initial search yielded 719 results; 171 articles were excluded as they were duplicates. Another 373 articles were excluded following the reading of titles and abstracts. Reading the full text of the 175 remaining articles, 68 were excluded because they did not meet our inclusion criteria. At the end of the selection, 107 articles were included in this systematic review (Figure 4). Details on the 107 selected items and their main contents such as number of cases, clinical information, diagnostic path, and therapeutic protocol are summarized in Supplementary Table S1. Through the evaluation of the 107 analyzed studies, 88 are classified as overall moderate RoB (MR), and 19 as serious overall RoB (SR). Definitely, in the analysis of 107 non-randomized studies, no article appears to have a critical RoB in the individual domain or in the analysis of the overall domain and, therefore, all the studies confirm evidence (Supplementary Table S2). A total number of 134 patients were described within the 107 selected studies, with a minimum age of 6 months old and a maximum age of 16 years (mean 6.06 ± 3.81). Only 31 out of the 134 patients (23.1%) were female, and 103 (76.9%) were male. A total of 69 (51.5%) patients were affected by neurodevelopmental disorders, and 41 (31.3%) suffered of autistic spectrum disorder. On the clinical examination, 66 out of 134 (49.3%) patients showed cutaneous manifestations of scurvy, 92 of 134 (68.7%) patients showed musculoskeletal manifestation and 112 out of 134 (83.6%) patients showed oral manifestation. Among these 112 patients with oral manifestations, 1 of 112 (0.89%) was affected by oral ulcers, 1 of 112 (0.89%) had a chronic glossitis and 110 of 112 (98.21%) had gingival involvement, such as gingival bleeding,

swelling and hypertrophy. The diagnosis was based on clinical hypothesis for 132 of the 134 patients (99.3%); serological dosage was obtained in 103 out of 134 patients (76.9%), histological examination was performed in 27 patients (20.1%); an ex juvantibus diagnostic approach was carried out in 17 cases (12.7%). The mean vitamin C serum level was 0.105 ± 0.118 mg/dl ranging from 0.000 to 0.700 mg/dl. Therapy was described in 78/134 patients. Only one patient of the 78 (1.28%) did not undergo vitamin C administration, but was only prescribed dietary recommendations. The remaining 77 patients (98.72%) underwent vitamin C therapy with a mean dosage of 384.81 ± 272.74 mg ranging from 50 to 1500 mg daily.

**Figure 4.** PRISMA flowchart.
