End-stage renal disease: a risk factor of deep neck infection - a nationwide follow-up study in Taiwan

Abstract

Background: Uremia is likely a risk factor for deep neck infection (DNI). However, only a few relevant cases have been reported, and evidence sufficient to support this hypothesis is lacking. The aim of the study is to investigate the effects of end-stage renal disease (ESRD) on DNI.

Methods: We used the database of the Registry for Catastrophic Illness Patients (RFCIP), a subset of the National Health Insurance Research Database (NHIRD) in Taiwan, to conduct a retrospective follow-up study. Between 1997 and 2013, a total of 157,340 patients in Taiwan with ESRD who received dialysis were registered in the RFCIP, whom were matched with a database consisting of 1,000,000 randomly selected patients who represented the national population, to conduct the follow-up study for investigating the incidence of DNI in the ESRD and control cohorts.

Results: In the ESRD group, 280 DNIs were identified with an incidence rate of 43 per 100,000 person-years. In the comparison group, 194 DNIs were identified with an incidence rate of 20 per 100,000 person-years. The incidence rate ratio was 2.16 (p < 0.001). Kaplan-Meier analysis indicated that the ESRD group had a significantly higher cumulative incidence of DNI (p < 0.001). According to Cox regression analysis, the hazard ratio of ESRD for DNI was 2.23 (p < 0.001). The therapeutic methods (non-surgery and surgery), performance of tracheostomy, duration of hospitalization did not differ significantly between the two groups, except more ESRD-DNI patients were admitted to intensive care units. The mortality rate of patients with DNI in the ESRD group was significantly higher than that in the control group (8.6% for ESRD vs 3.6% for control, p = 0.032). Furthermore, the Kaplan-Meier analysis demonstrated a poorer survival outcome in the ESRD group (p = 0.029). However, the individual survival outcomes following non-surgical and surgical therapies in the ESRD group did not differ significantly (p = 0.31).

Conclusions: ESRD is a predisposing factor for DNI, increasing its risk by twofold. In the patients with ESRD, DNI was not associated with higher rates of surgical debridement, tracheostomy, and mediastinal complications or longer hospital stays; however, it was associated with poorer survival outcomes, regardless of the therapeutic method.

Keywords: Abscess; Cellulitis; Cervical; Dialysis; ESRD; Failure; Kidney; NHIRD; Nephropathy; Predisposing.

Fig. 1

Enrolment schema of the study and comparison cohorts. Patients with ESRD were identified since 1997 to 2013 in Taiwan from RFCIP database, and a total of 157,340 cases were collected. 503 cases were excluded due to DNI occurrence prior to the index date of ESRD. A total of 156,837 cases with ESRD were eligible for a study cohort. LHID2000 database consisting of 989,480 insurants representing the general population in Taiwan was used to match with the study cohort for gender, age, urbanized, income level, DM and HTN with 1:1 fusion. Finally, there were 127,283 patients with ESRD (study group) and 127,283 patients without ESRD (control group) to conduct the follow-up study. The follow-up ended on DNI identified based on the ICD-9 codes of 528.3, 478.22, 478.24, and 682.11, death, or the end of 2013. Abbreviations: ESRD, end-stage renal disease; RFCIP, Registry for Catastrophic Illness Patients; LHID2000, Longitudinal Health Insurance Database 2000; DNI, deep neck infection; ICD-9, International Classification of Diseases, Ninth Revision; DM, diabetes mellitus; HTN, hypertension

Fig. 2

Cumulative incidence of DNI for ESRD versus non-ESRD. The Kaplan-Meier analysis demonstrated the cumulative DNI identified in the study and control cohorts, respectively, during the follow-up period (1997–2013). The log-rank test revealed a significantly higher cumulative incidence in the ESRD group (p < 0.001)

Fig. 3

Kaplan-Meier-estimated overall survival distributions for ESRD-DNI versus non-ESRD-DNI. The Kaplan-Meier analysis demonstrated the survival outcomes of the study and control cohorts, respectively, in the 3 months after DNI. The log-rank test revealed a significantly poorer survival result in the ESRD group (p = 0.029)

Fig. 4

Kaplan-Meier-estimated individual survival distributions from ESRD-DNI patients for surgery versus non-surgery. The Kaplan-Meier analysis demonstrated the individual survival outcomes of “surgery” and “non-surgery” therapies in the ESRD cohort in the 3 months after DNI, and the log-rank test revealed no significant difference between the two therapeutic methods for the survival results (p = 0.310)