Sequential Impact of Diabetes Mellitus on Deep Neck Infections: Comparison of the Clinical Characteristics of Patients with and without Diabetes Mellitus

Background: Deep neck infections (DNIs) can compromise the airway and are associated with high morbidity and mortality rates. Diabetes mellitus (DM) is a metabolic disorder characterized by chronic hyperglycemia that is associated with several comorbidities. We compared the clinical characteristics of DNI patients with and without DM. Methods: This study recorded the relevant clinical variables of 383 patients with DNIs between November 2016 and September 2022; of those patients, 147 (38.38%) had DM. The clinical factors between DNI patients with and without DM were assessed. Results: Patients with DM were older (p < 0.001), had higher white blood cell counts (p = 0.029) and C-reactive protein levels (CRP, p < 0.001), had a greater number of deep neck spaces (p = 0.002) compared to patients without DM, and had longer hospital stays (p < 0.001). Klebsiella pneumoniae was cultured more frequently from patients with DM than those without DM (p = 0.002). A higher CRP level (OR = 1.0094, 95% CI: 1.0047–1.0142, p < 0.001) was a significant independent risk factor for DM patients with prolonged hospitalization. The lengths of hospital stays in patients with poorly controlled DM were longer than those with well-controlled DM (p = 0.027). Conclusions: DNI disease severity and outcomes were worse in patients with DM than those without DM. Antibiotics effective against Klebsiella pneumoniae should be used for DNI patients with DM. DNI patients with DM and high CRP levels had more prolonged hospitalizations. Appropriate blood glucose control is essential for DNI patients with DM.


Introduction
A deep neck infection (DNI) is a lethal emergent bacterial infection of the deep neck spaces that can lead to cellulitis or abscess formation [1].Although DNIs usually result from upper aerodigestive tract infections, such as odontogenic infections and pharyngotonsillitis [2][3][4], there is an increasing proportion of DNI cases with an unknown primary origin [5].Treating DNIs is difficult because the infection is in close proximity to vital structures and the head and neck region has a complex anatomy [6].Despite the availability of antibiotics, advances in diagnostic technology, and surgical debridement, DNIs are still capable of causing serious complications, including airway obstruction [7], and can spread to the mediastinum [2].
Patients with comorbidities, such as alcohol consumption, drug abuse, and diabetes mellitus (DM), are at increased risk of complications from DNIs [8].DM affects almost 4% of the total population of Taiwan [9].It is also the most common comorbidity in DNIs [10].Animal and in vitro studies have found that hyperglycemia impairs the immune functions of hosts, including cellular immunity, complement activation, and neutrophil bactericidal function [11][12][13].In turn, immune deficiency predisposes DM patients to invasive infections [14].Multiple host factors in DM patients may increase the risk of skin and soft tissue infections, including uncontrolled hyperglycemia, disruption of the skin barrier, poor vascularization, and immune system dysfunction [15,16].Uncontrolled hyperglycemia is the main culprit behind infection.Furthermore, the clinical course of DNIs differs between patients with and without DM [17,18].The management of patients with a DNI and DM is complicated.In this study, we compared the clinical characteristics, outcomes, and prognosis of DNIs between patients with and without DM.

Materials and Methods
We retrospectively reviewed the medical records of 383 patients diagnosed with DNIs between November 2016 and September 2022 at the main branch of Chang Gung Memorial Hospital Linkou, a tertiary medical center in Taiwan.The patients were diagnosed on the basis of their clinical presentation and computed tomography (CT) findings (Figures 1 and 2) [19].The management of DNIs involved antibiotics and open surgical incision and drainage (I&D).Pathogen cultures were performed on specimens obtained using an aseptic procedure.Because primary culture results are often obtained after >3 days, empirical antibiotics were routinely administered prior to obtaining the culture results [20].Ceftriaxone 1 gm q12h and metronidazole 500 mg q8h were used as empirical antibiotics for aerobic and anaerobic bacteria [17,21,22].
Patients with comorbidities, such as alcohol consumption, drug abuse, and diabetes mellitus (DM), are at increased risk of complications from DNIs [8].DM affects almost 4% of the total population of Taiwan [9].It is also the most common comorbidity in DNIs [10].Animal and in vitro studies have found that hyperglycemia impairs the immune functions of hosts, including cellular immunity, complement activation, and neutrophil bactericidal function [11][12][13].In turn, immune deficiency predisposes DM patients to invasive infections [14].Multiple host factors in DM patients may increase the risk of skin and soft tissue infections, including uncontrolled hyperglycemia, disruption of the skin barrier, poor vascularization, and immune system dysfunction [15,16].Uncontrolled hyperglycemia is the main culprit behind infection.Furthermore, the clinical course of DNIs differs between patients with and without DM [17,18].The management of patients with a DNI and DM is complicated.In this study, we compared the clinical characteristics, outcomes, and prognosis of DNIs between patients with and without DM.

Materials and Methods
We retrospectively reviewed the medical records of 383 patients diagnosed with DNIs between November 2016 and September 2022 at the main branch of Chang Gung Memorial Hospital Linkou, a tertiary medical center in Taiwan.The patients were diagnosed on the basis of their clinical presentation and computed tomography (CT) findings (Figures 1 and 2) [19].The management of DNIs involved antibiotics and open surgical incision and drainage (I&D).Pathogen cultures were performed on specimens obtained using an aseptic procedure.Because primary culture results are often obtained after >3 days, empirical antibiotics were routinely administered prior to obtaining the culture results [20].Ceftriaxone 1 gm q12h and metronidazole 500 mg q8h were used as empirical antibiotics for aerobic and anaerobic bacteria [17,21,22].
There were 147 (38.38%) and 236 (61.62%) patients with and without DM, respectively.All patients with DM were treated with appropriate anti-diabetic medications.

Exclusion Criteria
Patients with previous head and neck surgery, radiotherapy to the head and neck region, or accidental swallowing of a foreign body were excluded.In this article, we only enrolled patients with type II DM and excluded patients with type I DM.In total, 383 patients with DNIs were included, including 147 with DM.

Data Collection
We collected data on gender, age, chief complaint duration, length of hospital stays, white blood cell (WBC) count, C-reactive protein (CRP) level, duration of DM, involvement of multiple spaces (≥3), surgical I&D, tracheostomy, deep neck space involvement, mediastinitis, pneumonia, necrotizing fasciitis, septic shock, and identified pathogens.

Ethics Statement
This study was approved by the Institutional Review Board (IRB) of Chang Gung Medical Foundation (IRB no.: 202201624B0).The sample size of the study was not determined based on any criteria.We included patients based on IRB-approved enrollment time.Data were collected retrospectively and anonymized before the analysis.The IRB waived the requirement for informed consent from patients.

Statistical Analysis
The Kolmogorov-Smirnov test showed that the data were not normally distributed.Therefore, the chi-square test and Mann-Whitney U test were used to compare categorical There were 147 (38.38%) and 236 (61.62%) patients with and without DM, respectively.All patients with DM were treated with appropriate anti-diabetic medications.

Exclusion Criteria
Patients with previous head and neck surgery, radiotherapy to the head and neck region, or accidental swallowing of a foreign body were excluded.In this article, we only enrolled patients with type II DM and excluded patients with type I DM.In total, 383 patients with DNIs were included, including 147 with DM.

Data Collection
We collected data on gender, age, chief complaint duration, length of hospital stays, white blood cell (WBC) count, C-reactive protein (CRP) level, duration of DM, involvement of multiple spaces (≥3), surgical I&D, tracheostomy, deep neck space involvement, mediastinitis, pneumonia, necrotizing fasciitis, septic shock, and identified pathogens.

Ethics Statement
This study was approved by the Institutional Review Board (IRB) of Chang Gung Medical Foundation (IRB no.: 202201624B0).The sample size of the study was not determined based on any criteria.We included patients based on IRB-approved enrollment time.Data were collected retrospectively and anonymized before the analysis.The IRB waived the requirement for informed consent from patients.

Statistical Analysis
The Kolmogorov-Smirnov test showed that the data were not normally distributed.Therefore, the chi-square test and Mann-Whitney U test were used to compare categorical and continuous variables, respectively.We performed univariate and multivariate analyses via logistic regression.We employed forward stepwise selection followed by multiple logistic regression; all variables included in the univariate analysis were entered into the final multivariable analysis.Data were analyzed using MedCalc software (version 18.6; MedCalc, Ostend, Belgium).p < 0.05 was considered to indicate statistical significance.Table 2 compares differences in the clinical characteristics of patients with and without DM.Among the individuals studied, there were 147 patients with DM and 236 patients without DM.In terms of age, there was a notable disparity between the two groups (p < 0.001); those with DM were considerably older, with a mean age of 56.28 ± 16.65 compared to 49.36 ± 19.77 for those without DM.When it comes to WBC and CRP, both levels were significantly higher in patients with DM (p = 0.029 and p < 0.001).The mean duration of DM was 3.93 ± 4.27 years, and mean HbA1c was 6.74 ± 0.51%.Regarding anti-diabetic drugs, 35 patients (23.81%) received insulin control, 85 patients (57.83%) used one anti-diabetic drug, and 27 patients (18.36%) used more than two anti-diabetic drugs.The proportion of more than three deep neck spaces involved in patients with DM is significantly higher than those without DM (38.09% vs. 23.31%,p = 0.002).CRP = C-reactive protein (normal range < 5mg/L); DM = diabetes mellitus; N = numbers; SD = standard deviation; WBC = white blood cell (normal range: 3900-10,600/µL); Significant differences with p < 0.05 are shown in bold.
Table 6 displays the comparison between DNI patients with poorly controlled (HbA1c ≥ 6.5) and well-controlled (HbA1c < 6.5) DM.The analysis indicated that the lengths of hospital stays in patients with poorly controlled DM were longer than those without DM (p = 0.027).
DM has a significant impact on the course of infections.In Taiwan, disease control using appropriate medications is achieved in 60% of DM patients; the remaining patients have uncontrolled disease [9].DM represents a group of physiological dysfunctions characterized by hyperglycemia due to insufficient insulin production, glucagon hypersecretion (type 1 DM), or directly from insulin resistance (type 2 DM) [34,35].Type 1 DM is a chronic autoimmune disease that affects about 1% of the population in developed countries [36].This immune response is induced and promoted by the interaction of environmental as well as genetic factors [37].In contrast, in type 2 diabetes, insulin resistance appears to be the primary cause of hyperglycemia (affecting around 8.5% of the adult population) [36].Patients with systemic diseases such as DM may experience the opportunistic advancement of seemingly minor infections due to compromised immune systems and vascular function [18,38,39].Huang et al. [1] found that the prevalence of DM was higher in DNI patients than in the general population of Taiwan.DM patients have a higher risk of infection and impaired cutaneous wound and soft tissue healing [40].Thus, DNI patients with DM have more severe and extensive inflammation [41,42].
In the present study, DNI patients with DM were older and had longer hospital stays, a higher WBC count and CRP level, and a greater number of involved deep neck spaces compared to DNI patients without DM.Patients with DNI and DM have more serious infections, a worse prognosis, and a higher rate of Klebsiella pneumoniae infection compared to patients with DNI alone.
In our research, patients with DM were older and had a longer hospital stay than patients without DM.Older patients with systemic diseases have impaired defense against pathogenic infections and a worse recovery rate, leading to prolonged hospital stays [10].DM is an expected risk depending on the age of the patients.It is a risk factor that must be controlled in every surgery.In addition, DM patients also had a higher WBC counts and CRP levels, as well as greater numbers of involved deep neck spaces compared to patients without DM, suggesting that the DM patients had more severe inflammation and worse outcomes [43].Sideris et al. [44] found that CRP level correlated with blood glucose level.Furthermore, the involvement of multiple deep neck spaces was a risk factor for the need for tracheostomy [45].Cheng et al. [46] also found that DM patients were at an increased risk of necrotizing fasciitis.Based on the findings that DM patients have more severe DNIs and poor outcomes, blood glucose levels should be monitored and appropriately controlled.
Timely open surgery is essential for the treatment of DNIs, and repeated surgical I&D is necessary in some patients [47].Therapeutic needle aspiration has also been performed in select cases [10,48].Although pus cultures obtained via needle aspiration or surgical drainage can aid the selection of appropriate antibiotics, empirical antibiotics should be administered before the culture results become available [49].
A DNI is typically caused by polymicrobial aerobic and anaerobic organisms [50].Studies from different countries and areas have reported different pathogenic causes of DNIs [21,51].Furthermore, the causative organisms of DNIs may change over time and might be influenced by the choice of antibiotics [20].
We found that DM patients had a higher rate of Klebsiella pneumoniae infection than patients without DM, in line with previous studies [10,18].One study found that Klebsiella pneumoniae is responsible for more than half of all DNI cases among DM patients [1].Klebsiella pneumoniae is a Gram-negative aerobe [20].The higher rate of identification of Klebsiella pneumoniae in DNI patients with DM may be because of oropharyngeal colonization by Gram-negative bacilli and a defective host defense, particularly phagocytosis [52,53].Although odontogenic infection is the most common etiology of DNIs [24,54,55], Huang et al. [49] found that DM patients had a high culture rate of Klebsiella pneumoniae, irrespective of the odontogenic or nonodontogenic origin of their DNI.Klebsiella pneumoniae is susceptible to cefazolin, gentamicin, and trimethoprim/sulfamethoxazole [21].However, clindamycin should not be administered alone because of its weak effect against Klebsiella pneumoniae [49].
Uncontrolled hyperglycemia correlates with a fatal outcome in DM patients with infectious diseases [56].Appropriate blood glucose control reduces the microangiopathic and macroangiopathic complications of DM [57,58].To achieve a favorable prognosis, the blood glucose level should be maintained below 200 mg/dL [1,10].For optimal empiric coverage, it is recommended to use either penicillin in combination with a β-lactamase inhibitor or a β-lactamase-resistant antibiotic combined with a drug highly effective against most anaerobes [59,60].Moreover, the addition of gentamicin is for effective Gram-negative coverage against Klebsiella pneumoniae, especially in DM patients, when it is resistant to clindamycin [55].

Study Limitations
This was a retrospective study performed at a single tertiary hospital.The treatments and management for patients with DNIs may have been relatively similar.When performing research in the future, patients could be enrolled from multiple hospitals to reduce such biases.Additionally, our study included only 383 cases.An increased number of samples can provide more comprehensive results.
Ideally, patients with infections should be switched from oral hypoglycemic agents to insulin.However, some patients still want to maintain their original oral hypoglycemic agents for treatment during hospitalization.As long as the patient's condition did not worsen, we kept the patient's original oral hypoglycemic agents.
Diabetes complications, such as diabetic retinopathy and peripheral neuropathy, can lead to balance impairment [61].However, this retrospective article mainly discusses the impact of DM on DNI patients with potential risk of mortality and does not document whether the patients suffered from diabetic balance or vertigo.The effect of DM on balance is an important topic worthy of future research.

Conclusions
Our data showed several characteristic features among DM patients with DNIs, including an older age, elevated levels of WBC and CRP, the involvement of multiple spaces, and extended hospital stays.Therefore, we recommend a more rigorous approach when dealing with DNIs in DM patients, emphasizing the importance of good glycemic control to prevent the occurrence of more severe complications.In addition, since the prevalence of Klebsiella pneumoniae is increased in patients with DM, the selection of empiric antibiotics should also take this into consideration.

Figure 1 .
Figure 1.(A) Axial view and (B) coronal view of a DNI patient with DM. (A) indicates that the infection spread from the parapharyngeal space to the parotid space and the submandibular space, encompassing infection across multiple deep neck spaces.(B) depicts infection in the coronal plane.Arrowhead: epiglottis; Double dotted arrow: involved spaces of DNI; P: parapharyngeal space; Pa: parotid space; S: submandibular space.(300 × 300 dpi).

Figure 1 .
Figure 1.(A) Axial view and (B) coronal view of a DNI patient with DM. (A) indicates that the infection spread from the parapharyngeal space to the parotid space and the submandibular space, encompassing infection across multiple deep neck spaces.(B) depicts infection in the coronal plane.Arrowhead: epiglottis; Double dotted arrow: involved spaces of DNI; P: parapharyngeal space; Pa: parotid space; S: submandibular space.(300 × 300 dpi).

Figure 2 .
Figure 2. The coronal view showed extensive infection in the retropharyngeal spaces with poor DM control.Arrow: compromised and deviated airway; Asterisk; multiloculated deep neck abscess; Double dotted arrow: involved spaces of DNI.(300 × 300 dpi).

Figure 2 .
Figure 2. The coronal view showed extensive infection in the retropharyngeal spaces with poor DM control.Arrow: compromised and deviated airway; Asterisk; multiloculated deep neck abscess; Double dotted arrow: involved spaces of DNI.(300 × 300 dpi).

Table 1 .
Clinical characteristics of the 383 patients with DNIs.

Table 2 .
Comparison of clinical characteristics between patients with and without DM.

Table 3 .
Comparison of management and morbidity between patients with and without DM.

Table 4 .
Comparison of pathogens between patients with and without DM.

Table 5 .
Univariate and multivariate analyses of 147 DM patients for prolonged hospitalization.

Table 6 .
Comparison between DNI patients with poorly controlled and well-controlled DM.