Letter to the Editor: Time-based evidence supporting early percutaneous catheter drainage in pyogenic liver abscess

Abstract

A recent study by Qiu et al demonstrated that early ultrasound-guided percutaneous catheter drainage (PCD) is safe and effective for treating pyogenic liver abscess (PLA). Here, we provide complementary evidence from a temporal perspective using data from a multicenter cohort study. We previously reported that abscess diameter influences clinical outcomes and hospital stay in patients with PLA. Using the same cohort, we conducted a subgroup analysis of 366 patients who underwent PCD at three tertiary centers in Korea. Patients who underwent PCD within 3 days of admission had significantly shorter hospital stays and a substantially greater reduction in high-sensitivity C-reactive protein levels after 1 week. In multivariate analysis, diabetes mellitus, lower albumin levels, and PCD performed more than 3 days after admission were independently associated with prolonged hospital stay. These findings complement the imaging-based evidence reported by Qiu et al, by providing temporal evidence that supports an early PCD strategy.

Key Words: Liver abscess; Pyogenic; Hospitalization; Percutaneous drainage; Early intervention; Clinical application

Core Tip: Among 366 patients who underwent percutaneous catheter drainage (PCD) for pyogenic liver abscess at three tertiary centers in Korea, early PCD performed within 3 days of admission was associated with shorter hospital stays and substantially greater reduction in high-sensitivity C-reactive protein levels after 1 week. Delayed PCD (> 3 days after admission), diabetes mellitus, and lower serum albumin levels were independent predictors of prolonged hospitalization. Together with the imaging-based findings reported by Qiu et al, these temporal data support early PCD as a beneficial management strategy for pyogenic liver abscess.

TO THE EDITOR

We read with great interest the recent article by Qiu et al[1] published in the World Journal of Gastrointestinal Surgery, in which the authors demonstrated that early ultrasound-guided percutaneous catheter drainage (PCD) is safe and effective for treating pyogenic liver abscess (PLA), even when the abscess is not fully liquefied. In their retrospective analysis of 143 patients, those in the non-liquefied group who underwent early PCD had significantly shorter hospital stays, faster resolution of fever, and more rapid abscess shrinkage than those in the liquefied group, without an increased risk of complications. These findings challenge the long-standing view that PCD should be deferred until liquefaction is confirmed on imaging. In this correspondence, we extend and reinterpret findings from a previously established multicenter Korean cohort and provide complementary evidence addressing the same clinical question using a temporal rather than imaging-based definition of early PCD.

MULTICENTER COHORT ANALYSIS

We previously reported in BMC Infectious Diseases that the maximum diameter of the liver abscess independently predicted prolonged hospital stay and in-hospital mortality among 648 patients with PLA across three tertiary centers in Korea[2]. Using the same cohort, we performed a subgroup analysis of the 366 patients who underwent PCD during hospitalization. Patients were stratified according to procedure timing into an early PCD group (within 3 days of admission; n = 269, 73.5%) and a delayed PCD group (after 3 days; n = 97, 26.5%). Baseline characteristics of the two groups are summarized in Table 1. Compared with the delayed group, patients who underwent early PCD were younger, more likely to have biliary disease as the underlying etiology, and had higher levels of inflammatory marker, including erythrocyte sedimentation rate and high-sensitivity C-reactive protein (hs-CRP). Regarding abscess characteristics, patients in the early PCD group were more likely to have single abscesses and had significantly larger maximum abscess diameters.

Table 1 Baseline clinical characteristics stratified by timing of percutaneous catheter drainage, mean ± SD/n (%).

Characteristics	PCD within 3 days (n = 269)	PCD after 3 days (n = 97)	P value
Age (years)	64.5 ± 14.9	68.3 ± 13.7	0.028
Male sex	173 (64.3)	57 (58.8)	0.397
Significant alcohol drinking	24 (8.9)	12 (12.4)	0.436
Underlying disease
Malignancy	36 (13.4)	19 (19.6)	0.193
Biliary disease	62 (23.0)	34 (35.1)	0.030
Diabetes mellitus	83 (30.9)	30 (30.9)	1.000
Hypertension	106 (39.4)	41 (42.3)	0.710
Chronic liver disease	17 (6.3)	1 (1.0)	0.073
Laboratory findings
WBC (mm3)	14.4 ± 6.7	14.1 ± 7.3	0.760
Platelet (× 1000/mm3)	288.3 ± 282.3	260.2 ± 165.9	0.295
ESR (mm/hour)	69.3 ± 30.4	60.4 ± 31.7	0.019
AST (IU/L)	89.7 ± 104.6	92.5 ± 80.3	0.790
ALT (IU/L)	81.4 ± 87.7	77.6 ± 64.3	0.655
Total bilirubin (mg/dL)	1.2 ± 1.0	1.5 ± 1.9	0.183
Albumin (g/dL)	3.3 ± 0.5	3.4 ± 0.5	0.126
Creatinine (mg/dL)	1.0 ± 0.8	1.4 ± 1.4	0.015
hs-CRP (mg/L)	180.4 ± 80.7	154.8 ± 87.9	0.009
Procalcitonin (ng/mL)	16.4 ± 31.1	22.0 ± 27.4	0.245
Characteristics of abscess
Number (single vs multi, %)	85 (31.6)	18 (18.6)	0.021
Maximal diameter (cm)	6.2 ± 2.8	5.6 ± 2.2	0.035

PCD: Percutaneous catheter drainage; WBC: White blood cells; ESR: Erythrocyte sedimentation rate; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; hs-CRP: High-sensitivity C-reactive protein.

EARLY PCD IS ASSOCIATED WITH FASTER INFLAMMATORY RESOLUTION AND SHORTER HOSPITAL STAY

Hematologic and biochemical parameters were reassessed 1 week after hospital admission (Table 2). At the 1-week follow-up, the early PCD group had significantly lower hs-CRP levels than the delayed PCD group (72.6 ± 57.2 mg/L vs 95.5 ± 68.5 mg/L, P = 0.004). Compared with baseline, hs-CRP levels decreased by 107.8 mg/L in the early PCD group compared with 59.3 mg/L in the delayed PCD group over the first week. Consistent with these findings, the early PCD group also had a significantly shorter hospital stay than the delayed PCD group (19.2 ± 12.1 days vs 22.5 ± 11.6 days, P = 0.022). In contrast, in-hospital mortality did not differ significantly between the two groups (1.9% vs 1.0%, P = 0.933).

Table 2 Hematologic and biochemical parameters 1 week after treatment, mean ± SD.

Laboratory parameter	PCD within 3 days (n = 269)	PCD after 3 days (n = 97)	P value
WBC (mm3)	11.4 ± 5.1	12.3 ± 5.4	0.136
Hemoglobin (g/dL)	11.4 ± 1.6	11.1 ± 1.8	0.148
Platelet (× 1000/mm3)	384.7 ± 157.2	321.6 ± 157.5	0.001
ESR (mm/hour)	67.6 ± 24.6	63.5 ± 28.8	0.272
AST (IU/L)	37.5 ± 41.3	40.5 ± 38.0	0.539
ALT (IU/L)	37.2 ± 36.0	40.1 ± 29.6	0.441
Total bilirubin (mg/dL)	0.8 ± 1.1	1.0 ± 1.2	0.043
Albumin (g/dL)	3.1 ± 0.5	3.1 ± 0.5	0.734
Creatinine (mg/dL)	0.7 ± 0.8	0.9 ± 1.2	0.174
hs-CRP (mg/L)	72.6 ± 57.2	95.5 ± 68.5	0.004
Procalcitonin (ng/mL)	2.0 ± 6.5	5.2 ± 14.2	0.327

PCD: Percutaneous catheter drainage; WBC: White blood cells; ESR: Erythrocyte sedimentation rate; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; hs-CRP: High-sensitivity C-reactive protein.

INDEPENDENT PROTECTIVE EFFECT IN MULTIVARIABLE ANALYSIS

To identify factors independently associated with prolonged hospital stay (> 14 days), we performed a multivariable logistic regression analysis (Table 3). Along with diabetes mellitus and lower serum albumin levels, PCD performed within 3 days of hospital admission was independently associated with a reduced risk of prolonged hospital stay (odds ratio = 0.33; P = 0.001), indicating an independent protective effect.

Table 3 Factors associated with prolonged hospital stay (> 14 days).

Variable	Univariable analysis		Multivariable analysis
	P value	OR (95%CI)	P value	OR (95%CI)
Age	0.178	1.01 (0.99-1.03)
Male sex	0.340	0.78 (0.47-1.28)
Malignancy	0.013	3.07 (1.36-8.24)	0.096	2.19 (0.93-6.07)
Biliary disease	0.811	1.07 (0.63-1.87)
Diabetes mellitus	0.036	1.81 (1.06-3.23)	0.033	1.88 (1.07-3.42)
Hypertension	0.263	1.33 (0.81-2.19)
Altered mentality at admission	0.410	0.54 (0.13-2.69)
Shock at admission	0.657	1.34 (0.41-5.97)
WBC (mm3)	0.246	1.02 (0.99-1.06)
Hemoglobin (g/dL)	0.055	0.88 (0.77-1.00)
ALT (IU/L)	0.485	1.00 (1.00-1.00)
Total bilirubin (mg/dL)	0.307	1.13 (0.92-1.49)
Albumin (g/dL)	< 0.001	0.38 (0.23-0.61)	0.003	0.44 (0.25-0.74)
Creatinine (mg/dL)	0.229	1.23 (0.93-1.86)
hs-CRP (mg/L)	0.268	1.00 (1.00-1.00)
Number of abscess	0.870	1.05 (0.62-1.80)
Maximal abscess diameter (cm)	0.015	1.13 (1.03-1.25)	0.079	1.10 (0.99-1.24)
Early PCD (within 3 days of admission)	0.003	0.38 (0.19-0.70)	0.001	0.33 (0.16-0.63)

Bold rows indicate factors that remained statistically significant in multivariable analysis. WBC: White blood cells; ALT: Alanine aminotransferase; PCD: Percutaneous catheter drainage; hs-CRP: High-sensitivity C-reactive protein; OR: Odds ratio; CI: Confidence interval.

TIME AND IMAGING-BASED DEFINITIONS OF EARLY PCD ARE COMPLEMENTARY

The optimal timing of drainage for PLA remains an unresolved, yet clinically important question[3]. Numerous studies have evaluated the relative effectiveness of PCD and percutaneous needle aspiration according to abscess size, abscess number, and timing of intervention[4-7]. Patient-specific clinical factors should also be considered when selecting an appropriate drainage strategy[8,9]. Recent meta-analyses have further refined the understanding of the role of PCD[10]. The study by Qiu et al[1] contributes to this growing body of evidence by demonstrating the benefit of early intervention.

Our findings support and extend those of Qiu et al[1] by providing a complementary temporal perspective. The authors defined early PCD as the absence of complete liquefaction on ultrasound, whereas we defined it as the elapsed time from admission. In our earlier study, the timing of PCD was examined in an overall cohort of patients with PLA, in which the maximum abscess diameter, rather than early PCD, emerged as the strongest predictor of clinical outcomes. However, this analysis included all patients with PLA regardless of whether they underwent PCD. In the present subgroup analysis, which was restricted to patients who underwent PCD, the timing of PCD emerged as an independent prognostic factor. Taken together with the findings of Qiu et al[1], our results highlight the importance of the timing of the procedure, regardless of how early it is defined operationally. In clinical practice, these two definitions represent sequential decision points: Imaging determines whether the lesion is suitable for drainage, whereas timing determines how promptly drainage should be performed once the lesion is deemed drainable.

CONCLUSION

Taken together, the imaging-based evidence reported by Qiu et al[1] and our temporal multicenter subgroup analysis converge on a single clinical message: Early PCD should be considered for patients with PLA who have a clearly identifiable drainable lesion, rather than delaying intervention through watchful waiting until liquefaction is confirmed on imaging. Markers of disease severity, such as abscess diameter and serum albumin level, may help identify patients at increased risk of prolonged hospital stay; however, these factors are not modifiable. In contrast, the timing of drainage is a modifiable factor that is directly under the clinician’s control. Nonetheless, our findings should be interpreted with caution, as they are derived from a retrospective, post hoc subgroup analysis of a previously established multicenter cohort. Residual confounding and selection bias cannot be excluded, and the observed associations do not establish causality. Prospective multicenter studies that integrate imaging- and time-based definitions of early PCD are warranted to refine clinical practice guidelines and identify patient subgroups that derive the greatest benefit from early intervention.

ACKNOWLEDGEMENTS

The cohort data analyzed in this study were originally collected for our previously published multicenter study (Lee et al[2]). We wish to thank the co-investigators of the previous study, for their contributions to the original cohort assembly.