Scholarlink

Background <p>Malnutrition affects 50–90% of children with cancer in low- and middle-income countries (LMICs) and worsens clinical outcomes. The Nutrition Screening Tool for Childhood Cancer (SCAN) is cancer-specific but has not been validated in Brazilian Portuguese. We aimed to evaluate the diagnostic accuracy of the Brazilian Portuguese SCAN (SCAN pt-BR) against STRONGkids and to examine its association with prolonged hospitalization and ICU admission.</p> Methods <p>Prospective diagnostic-accuracy study at a tertiary pediatric oncology center in Brazil (February–December 2021). Consecutive inpatients aged 1–18 years with expected stay ≥48 h were screened within 24–48 h of admission by pediatric dietitians using SCAN pt-BR and STRONGkids.</p> Results <p>Of 262 children (median age 8.8 years, 52.7% male, 41.2% hematologic malignancies), 70.2% were classified as at risk by SCAN pt-BR. Against STRONGkids ≥4, SCAN showed sensitivity 94.6%, specificity 43.2%, and ROC-AUC 0.863. SCAN risk independently predicted prolonged stay (adjusted OR 2.33, 95% CI 1.29–4.30, <i>p</i> = 0.006) but not ICU admission (adjusted OR 0.96, <i>p</i> = 0.899).</p> Conclusion <p>SCAN pt-BR showed adequate screening performance and independently predicted prolonged hospitalization in Brazilian children with cancer. These results support its use for admission nutritional triage in LMIC pediatric oncology settings.</p> Impact <p><UnorderedList Mark="Bullet"> <ItemContent> <p>SCAN identified 70.2% of children as at nutritional risk and predicted prolonged hospitalization (adjusted OR 2.33, 95% CI 1.29–4.30) independently of age, tumor group, and STRONGkids classification.</p> </ItemContent> <ItemContent> <p>Against STRONGkids high-risk classification, SCAN pt-BR showed high sensitivity (94.6%), high negative predictive value (93.6%), and ROC-AUC of 0.863.</p> </ItemContent> <ItemContent> <p>Gastrointestinal symptoms, poor oral intake, and recent weight loss were the most frequently endorsed symptom-related items that children classified as at nutritional risk, suggesting that SCAN detects acute nutritional deterioration rather than static disease characteristics.</p> </ItemContent> <ItemContent> <p>Routine application at admission could direct early dietetic referral to children most likely to require prolonged inpatient care.</p> </ItemContent> </UnorderedList></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Nutritional screening predicts prolonged stay in LMIC pediatric oncology settings: SCAN validation in Brazilian children

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月15日

查看全文
  • Mariana Murra Duarte,
  • Marielle Ferreira da Silva,
  • Maynara Cristina Alexandre,
  • Marise Lopes Fermino,
  • Ana Laura Balduino do Nascimento,
  • Wilson Elias de Oliveira Jr

摘要

Background

Malnutrition affects 50–90% of children with cancer in low- and middle-income countries (LMICs) and worsens clinical outcomes. The Nutrition Screening Tool for Childhood Cancer (SCAN) is cancer-specific but has not been validated in Brazilian Portuguese. We aimed to evaluate the diagnostic accuracy of the Brazilian Portuguese SCAN (SCAN pt-BR) against STRONGkids and to examine its association with prolonged hospitalization and ICU admission.

Methods

Prospective diagnostic-accuracy study at a tertiary pediatric oncology center in Brazil (February–December 2021). Consecutive inpatients aged 1–18 years with expected stay ≥48 h were screened within 24–48 h of admission by pediatric dietitians using SCAN pt-BR and STRONGkids.

Results

Of 262 children (median age 8.8 years, 52.7% male, 41.2% hematologic malignancies), 70.2% were classified as at risk by SCAN pt-BR. Against STRONGkids ≥4, SCAN showed sensitivity 94.6%, specificity 43.2%, and ROC-AUC 0.863. SCAN risk independently predicted prolonged stay (adjusted OR 2.33, 95% CI 1.29–4.30, p = 0.006) but not ICU admission (adjusted OR 0.96, p = 0.899).

Conclusion

SCAN pt-BR showed adequate screening performance and independently predicted prolonged hospitalization in Brazilian children with cancer. These results support its use for admission nutritional triage in LMIC pediatric oncology settings.

Impact

SCAN identified 70.2% of children as at nutritional risk and predicted prolonged hospitalization (adjusted OR 2.33, 95% CI 1.29–4.30) independently of age, tumor group, and STRONGkids classification.

Against STRONGkids high-risk classification, SCAN pt-BR showed high sensitivity (94.6%), high negative predictive value (93.6%), and ROC-AUC of 0.863.

Gastrointestinal symptoms, poor oral intake, and recent weight loss were the most frequently endorsed symptom-related items that children classified as at nutritional risk, suggesting that SCAN detects acute nutritional deterioration rather than static disease characteristics.

Routine application at admission could direct early dietetic referral to children most likely to require prolonged inpatient care.