# 人工智能未能提醒接受CT碘对比剂检查且eGFR<30 mL/min/1.73m²患者的对比剂肾病风险

- **ID:** `medical/contrast-induced-nephropathy-missed`
- **领域:** medical
- **类别:** data_error
- **错误码:** `CIN-RISK-01`
- **验证级别:** ai_generated
- **修复率:** 82%

## 根因

AI模型在对比剂给药前遗漏了强制性肾功能检查，忽略了KDIGO指南要求在7天内评估eGFR，并对eGFR<30的患者采用替代影像学或预防措施。

## 版本兼容性

| 版本 | 状态 | 引入 | 弃用 |
|------|------|------|------|
| Radiology Information System v5.2.1 | active | — | — |
| Epic Radiant 2024 | active | — | — |
| DICOM Conformance Statement 3.0 | active | — | — |

## 解决方案

1. ```
   Check eGFR within 7 days prior to contrast; if eGFR <30, order non-contrast MRI or ultrasound instead of CT. Document in EHR: 'CIN risk - alternative imaging selected.'
   ```
2. ```
   If contrast is unavoidable (e.g., for acute pulmonary embolism), pre-administer 0.9% saline 1 mL/kg/h for 12 hours, use iso-osmolar contrast (iodixanol), and schedule post-procedure hemodialysis if eGFR <15. Implement in order set: 'CONTRAST-CIN-PROTOCOL'.
   ```
3. ```
   Integrate a CDS (Clinical Decision Support) rule into the CPOE system: if contrast order AND eGFR <30, block order and display: 'eGFR <30: CIN risk. Alternative imaging required. Contact radiology.'
   ```

## 无效尝试

- **** — In patients with severe renal impairment (eGFR <30), standard saline hydration is insufficient to prevent CIN without withholding contrast entirely or using iso-osmolar contrast agents. (65% 失败率)
- **** — Multiple large RCTs (e.g., ACT trial) have shown N-acetylcysteine provides no benefit over placebo for preventing CIN in high-risk patients. (90% 失败率)
- **** — Even low-volume contrast can cause nephropathy in eGFR <30 patients; the only safe approach is alternative imaging (MRI, ultrasound) or hemodialysis planning. (70% 失败率)