Hematologic and inflammatory parameters for determining severity of odontogenic infections at admission: a retrospective study

Abstract

Background: Severe odontogenic infections in the head and neck region, especially necrotizing soft tissue infection (NSTI) and deep neck abscess, are potentially fatal due to their delayed diagnosis and treatment. Clinically, it is often difficult to distinguish NSTI and deep neck abscess in its early stage from cellulitis, and the decision to perform contrast-enhanced computed tomography imaging for detection is often a challenge. This retrospective case-control study aimed to examine the utility of routine blood tests as an adjunctive diagnostic tool for NSTI in the head and neck region and deep neck abscesses.

Methods: Patients with severe odontogenic infections in the head and neck region that required hospitalization were classified into four groups. At admission, hematologic and inflammatory parameters were calculated according to the blood test results. In addition, a decision tree analysis was performed to detect NSTI and deep neck abscesses.

Results: There were 271 patients, 45.4% in Group I (cellulitis), 22.5% in Group II (cellulitis with shallow abscess formation), 27.3% in Group III (deep neck abscess), and 4.8% in Group IV (NSTI). All hematologic and inflammatory parameters were higher in Groups III and IV. The Laboratory Risk Indicator for Necrotizing Fasciitis score, with a cut-off value of 6 and C-reactive protein (CRP) + the neutrophil-to-lymphocyte ratio (NLR), with a cut-off of 27, were remarkably useful for the exclusion diagnosis for Group IV. The decision tree analysis showed that the systemic immune-inflammation index (SII) of ≥ 282 or < 282 but with a CRP + NLR of ≥ 25 suggests Group III + IV and the classification accuracy was 89.3%.

Conclusions: Hematologic and inflammatory parameters calculated using routine blood tests can be helpful as an adjunctive diagnostic tool in the early diagnosis of potentially fatal odontogenic infections. An SII of ≥ 282 or < 282 but with a CRP + NLR of ≥ 25 can be useful in the decision-making for performing contrast-enhanced computed tomography imaging.

Keywords: Cellulitis; Contrast-enhanced computed tomography; Deep neck abscess; Necrotizing soft tissue infection; Neutrophil-to-lymphocyte ratio; Systemic immune-inflammation index.

Fig. 1

Flowchart of data collection and cleaning

Fig. 2

ROC analysis for the diagnostic performance of each of the hematologic and inflammatory parameters for NSTI. ROC receiver operating characteristic, NTSI necrotizing soft tissue infection, LRINEC laboratory risk indicator for necrotizing fasciitis, CRP C-reactive protein, WBC White blood cell, NLR Neutrophil-to-lymphocyte ratio, PLR Platelet-to-lymphocyte ratio, SII systemic immune-inflammation index

Fig. 3

Decision tree analysis for discrimination between Groups I + II and Groups III + IV. The node (the frame) shows the group (I + II, III + IV), the number of cases, as well as the proportion of the total data. If the element (split condition expression) is satisfied, it goes left (node); if not, to the right. The color of the node indicates “I + II > III + IV” for blue and “III + IV > I + II” for green. The darker the node’s color, the lower its entropy (average information content); that is, the higher its purity. For example, if a node is dark green, it indicates that the node is mostly III + IV, indicating high classification accuracy. Note that the one order of magnitude for SII is smaller than that of the original. Group I: cellulitis; Group II: cellulitis with shallow abscess formation; Group III: deep neck abscess; and Group IV: necrotizing soft tissue infection. CRP C-reactive protein, NLR neutrophil-to-lymphocyte ratio, PLR platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index