Background <p>Over the past decades, the widespread use of advanced imaging has improved detection of psoas abscess (PA), yet evidence from low- and middle-income settings remains limited. We aimed to describe the clinical, etiological and microbiological characteristics of PA and to explore clinically relevant factors associated with in-hospital mortality using data from Vietnam.</p> Methods <p>We retrospectively reviewed all inpatients with CT/MRI-confirmed PA at Cho Ray Hospital, Vietnam (January 2020–December 2024). Demographics, comorbidities, presentation, laboratory/imaging findings, microbiology, treatments, and outcomes were extracted. Sepsis/septic shock followed Sepsis-3 definitions.</p> Results <p>Among 233 patients (mean age 52.3 ± 17.5 years; 59.7% male), diabetes was the most common comorbidity (28.8%). Flank pain (55.8%) and fever (39.1%) predominated. Primary PA accounted for 68.2%; spinal sources were most frequent (24.5%) among secondary cases. Cultures were positive in 38.7%. <i>Staphylococcus aureus</i> (41.4%) and <i>Escherichia coli</i> (34.3%) were predominant. Sepsis/septic shock developed in 9.0%, and in-hospital mortality was 5.2%. In exploratory univariable Cox analysis, several clinical variables, including age ≥ 65 years, tachycardia, tachypnea, hypotension, thrombocytopenia, and treatment modality, were associated with in-hospital mortality. Among these, tachypnea and thrombocytopenia appeared particularly clinically relevant as simple bedside indicators of systemic deterioration.</p> Conclusions <p>In this large Vietnamese cohort, PA chiefly affected middle-aged and older adults and often presented nonspecifically. In exploratory analysis, tachypnea and thrombocytopenia were identified as clinically relevant bedside markers associated with in-hospital mortality; these findings should be interpreted as hypothesis-generating rather than confirmatory.</p> Trial registration <p>Not applicable (retrospective study).</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

Clinical characteristics and factors associated with in-hospital mortality in psoas abscess: a retrospective cohort study from Vietnam

  • Kinh Luan Thai,
  • Ngoc Huy Pham,
  • Thanh Tuan Nguyen,
  • Nguyen Tran-Ngoc,
  • Xuan Thai Ngo

摘要

Background

Over the past decades, the widespread use of advanced imaging has improved detection of psoas abscess (PA), yet evidence from low- and middle-income settings remains limited. We aimed to describe the clinical, etiological and microbiological characteristics of PA and to explore clinically relevant factors associated with in-hospital mortality using data from Vietnam.

Methods

We retrospectively reviewed all inpatients with CT/MRI-confirmed PA at Cho Ray Hospital, Vietnam (January 2020–December 2024). Demographics, comorbidities, presentation, laboratory/imaging findings, microbiology, treatments, and outcomes were extracted. Sepsis/septic shock followed Sepsis-3 definitions.

Results

Among 233 patients (mean age 52.3 ± 17.5 years; 59.7% male), diabetes was the most common comorbidity (28.8%). Flank pain (55.8%) and fever (39.1%) predominated. Primary PA accounted for 68.2%; spinal sources were most frequent (24.5%) among secondary cases. Cultures were positive in 38.7%. Staphylococcus aureus (41.4%) and Escherichia coli (34.3%) were predominant. Sepsis/septic shock developed in 9.0%, and in-hospital mortality was 5.2%. In exploratory univariable Cox analysis, several clinical variables, including age ≥ 65 years, tachycardia, tachypnea, hypotension, thrombocytopenia, and treatment modality, were associated with in-hospital mortality. Among these, tachypnea and thrombocytopenia appeared particularly clinically relevant as simple bedside indicators of systemic deterioration.

Conclusions

In this large Vietnamese cohort, PA chiefly affected middle-aged and older adults and often presented nonspecifically. In exploratory analysis, tachypnea and thrombocytopenia were identified as clinically relevant bedside markers associated with in-hospital mortality; these findings should be interpreted as hypothesis-generating rather than confirmatory.

Trial registration

Not applicable (retrospective study).