The Role of C-Reactive Protein and Neutrophil to Lymphocyte Ratio in Predicting the Severity of Odontogenic Infections in Adult Patients
C-反應蛋白和嗜中性白血球與淋巴球比率在預測成人患者齒源性感染嚴重程度中的作用

2.1. Study Design and Ethics
2.1. 研究設計與倫理

The study was designed as a retrospective cohort of patients admitted for odontogenic infections to the Maxillofacial Surgery Department of City Emergency Hospital Timisoara (SCMUT), affiliated with the Victor Babes University of Medicine and Pharmacy from Timisoara between January 2017 to April 2022. These data were collected from digital and paper records only with the patient’s agreement and the ethical approval obtained from the Ethics Committee of SCMUT with the approval number I-27098 from 14 October 2022.
本研究設計為一項回顧性队列研究，研究對象為 2017 年 1 月至 2022 年 4 月期間因牙源性感染入院至蒂米什瓦拉市急診醫院（SCMUT）口腔頜面外科，該醫院隸屬於蒂米什瓦拉 Victor Babes 醫學與藥學大學。這些數據僅在徵得患者同意並獲得 SCMUT 倫理委員會的倫理批准（批准號 I-27098，日期為 2022 年 10 月 14 日）後，才從數位和紙質記錄中收集。

2.2. Patient Selection Process
2.2. 病患選擇流程

Patients above the age of eighteen were enrolled in the research. Infections of odontogenic origin were examined for inclusion according to the international classification of diseases (ICD-10) disease classification [31]. Patients whose medical records were incomplete were excluded from the research. Patients under the age of 18, pregnant women, and those with malignancy, immunodeficiency, or infections of origin other than odontogenic were excluded from the research, to avoid potential outliers in the levels of serum inflammatory markers. According to the Symptom Severity score (SS) presented in Table 1, eligible cases were divided into two groups based on the severity of the infection. The low-severity infection group consisted of mild to moderate infections, whereas the high-severity infection group included moderate to severe infections. At admission, the SS score of odontogenic infection created by Sainuddin et al. [30] and used in this study was calculated. Sepsis was defined by the recent guidelines in accordance with the sequential sepsis-related organ failure assessment score (SOFA) [32].
參與研究的對象為年滿十八歲的患者。根據國際疾病分類（ICD-10）疾病分類 [31]，檢查牙源性感染以納入研究。醫療記錄不完整的患者被排除在研究之外。為了避免血清發炎指標水平出現潛在的異常值，研究排除了 18 歲以下的患者、孕婦以及患有惡性腫瘤、免疫缺陷或非牙源性感染的患者。根據表 1 中呈現的症狀嚴重程度評分（SS），符合條件的病例根據感染的嚴重程度分為兩組。低嚴重程度感染組包括輕度至中度感染，而高嚴重程度感染組包括中度至重度感染。入院時，計算了 Sainuddin 等人 [30] 創建並在本研究中使用的牙源性感染 SS 評分。敗血症的定義符合近期指南，依據連續性敗血症相關器官衰竭評估評分（SOFA）[32]。

A convenience sampling method was employed to calculate the appropriate sample size. Considering the incidence of OI in the general population ranges between 0.05% and 0.1% [33,34], the computed ideal sample size was 34 patients, using a 99% confidence level and a 1 margin of error. Between January 2017 and April 2022, a total of 141 eligible patients identified with odontogenic infections were hospitalized at the Maxillofacial Surgery Department of the SCMUT. After deleting missing data and filtering by severity scores, 108 patients were eventually matched 1:1 by severity index and included in the study. The records were subsequently divided into two groups based on the primary anatomic space involved and the SS score: Group A consists of 54 individuals with a lower severity (SS score from 0 to 8 points); Group B consists of 54 patients with a greater severity (SS score from 9 to 16 points).
採用便利抽樣方法計算合適的樣本量。 考慮到一般人群中齒源性感染的發病率在 0.05% 到 0.1% 之間 [ 33, 34]，計算出的理想樣本量為 34 名患者，使用 99% 的置信水平和 1 的誤差範圍。 2017 年 1 月至 2022 年 4 月期間，共有 141 名符合條件且被診斷患有齒源性感染的患者在 SCMUT 的頜面外科住院。 在刪除缺失數據並按嚴重程度評分進行篩選後，最終將 108 名患者按嚴重程度指數 1:1 進行匹配，並納入研究。 隨後，根據主要受累的解剖空間和 SS 評分將記錄分為兩組：A 組由 54 名嚴重程度較低的個體組成（SS 評分為 0 到 8 分）； B 組由 54 名嚴重程度較高的患者組成（SS 評分為 9 到 16 分）。

2.3. Data Collection and Variables
2.3. 數據收集和變量

Demographic data and the patient’s medical history were collected. The hospital information system obtained the patients’ discharge reports, clinical evaluations, laboratory values, and imaging tests. Furthermore, routine blood samples, white blood cell count (WBC), hemogram indexes such as neutrophil and lymphocyte count, Neutrophil to Lymphocyte Ratio (NLR), and platelet count were evaluated. The variables considered for analysis comprised demographic data: age, gender, and place of origin, clinical presentation features (body temperature, trismus (mild, moderate, or severe), odontalgia (visual analog scale), mandibular pain (visual analog scale), dysfunctional disturbances of the masticatory system (mandibular dysfunction, headache, and unilateral chewing side)), edema, signs of obstruction (dyspnea, dysphagia), and signs of systemic infection (temperature >38.3 °C or <35.3 °C, heart rate > 90 bpm, respiratory rate > 20/min, blood pressure and WBC < 4 or >12 × 10³/μL) [35]. Routine blood sample on admission to the hospital: complete blood count, C-reactive protein, erythrocyte sedimentation rate (ESR), blood glucose levels, sodium and potassium, creatinine, and the glomerular filtration rate, blood urea nitrogen (BUN), aspartate transaminase (AST), alanine transaminase (ALT), clotting time, and swab culture with antibiogram. Research variables for serum parameters included the Neutrophil to Lymphocytes Ratio (NLR) obtained by dividing absolute Neutrophil and Lymphocyte counts.
收集了人口統計數據和患者的病史。 醫院信息系統獲取了患者的出院報告、臨床評估、實驗室數值和影像學檢查。 此外，還評估了常規血樣、白細胞計數 (WBC)、血象指標（如嗜中性白血球和淋巴球計數）、嗜中性白血球與淋巴球比率 (NLR) 和血小板計數。 分析考慮的變量包括人口統計數據：年齡、性別和原籍地、臨床表現特徵（體溫、牙關緊閉（輕度、中度或重度）、牙痛（視覺模擬量表）、下頜疼痛（視覺模擬量表）、咀嚼系統功能障礙（下頜功能障礙、頭痛和單側咀嚼側））、水腫、阻塞徵象（呼吸困難、吞嚥困難）和全身感染徵象（體溫 >38.3 °C 或 <35.3 °C、心率 > 90 bpm、呼吸頻率 > 20 次/分鐘、血壓和 WBC < 4 或 >12 × 10³/μL）[ 35]。 入院時的常規血樣：全血細胞計數、C-反應蛋白、紅血球沉降率 (ESR)、血糖水平、鈉和鉀、肌酐和腎絲球過濾率、血尿素氮 (BUN)、天門冬胺酸轉胺酶 (AST)、丙胺酸轉胺酶 (ALT)、凝血時間和帶抗生素敏感試驗的拭子培養。 血清參數的研究變量包括通過將絕對嗜中性白血球和淋巴球計數相除獲得的嗜中性白血球與淋巴球比率 (NLR)。

2.4. Statistical Analysis
2.4. 統計分析

Data were obtained electronically and deidentified. Mean values and standard deviations (SD), p-values, and correlation coefficient “r” of the laboratory values were calculated using the statistical analysis software MedCalc (MedCalc Software bv, Ostend, Belgium). Variables were compared between group A and group B, including the laboratory tests mentioned above related to the Severity Score (SS) of odontogenic infections. The Mann−Whitney U test was applied to compare non-normally distributed means, while Student’s t-test was used to compare normally-distributed data. Chi-square and Fischer’s exact tests were applied to verify a possible difference between the two groups regarding variables described as proportionate values. Logistic regression analysis was applied to determine the association between CRP and NLR. The hazard ratio and adjusted odds ratios were determined for the assessment of CRP and NLR as predictors for infection severity (represented by SS score severity). The area under the curve (AUC) was plotted for CRP and NLR to determine their accuracy in predicting the severity of odontogenic infections. A p-value < 0.05 was considered statistically significant when comparing the study variables.
數據以電子方式獲取並去識別化。使用統計分析軟件 MedCalc（MedCalc Software bv，奧斯滕德，比利時）計算實驗室值的平均值和標準差 (SD)、p 值和相關係數“r”。比較 A 組和 B 組之間的變量，包括與牙源性感染嚴重程度評分 (SS) 相關的上述實驗室測試。 Mann−Whitney U 檢驗用於比較非正態分佈的均值，而 Student's t 檢驗用於比較正態分佈的數據。應用卡方檢驗和 Fischer 精確檢驗來驗證兩組之間在描述為比例值的變量方面是否存在可能的差異。應用邏輯回歸分析來確定 CRP 和 NLR 之間的關聯。確定風險比和調整後的勝算比，以評估 CRP 和 NLR 作為感染嚴重程度（以 SS 評分嚴重程度表示）的預測指標。繪製 CRP 和 NLR 的曲線下面積 (AUC)，以確定它們在預測牙源性感染嚴重程度方面的準確性。當比較研究變量時，p 值 < 0.05 被認為具有統計學意義。

3.1. Demographic Characteristics of the Study Population
3.1. 研究人群的人口統計學特徵

In total, 544 patients diagnosed clinically and radiologically with odontogenic infections were admitted and hospitalized at the Maxillofacial Surgery Department, SCMUT, Romania, between January 2017 and April 2022. Only 108 patients met the inclusion criteria and were enrolled in the study, as described in Figure 1. The patients were further subcategorized according to the SS score into two groups as follows: Group A—the low-severity infection group with 54 patients whose severity score ranges from 0 to 8 points on the SS scale; Group B—the high-severity infection group including 54 patients with a severity score between 9 and over 16 points. Table 2 describes the comparison of background characteristics among patients with odontogenic infections. It was observed that men were more frequently involved with OI (55.6% in Group A and 66.7% in Group B). The mean age was 46.7 years in Group A (age range 18–81), compared with 51.7 years in Group B (age range 20–85), without a statistically significant difference (p-value = 0.150). However, the place of origin was significantly different between the study groups, with patients with more severe infections coming more frequently from rural regions (68.5% vs. 46.3%, p-value = 0.019). Additionally, patients with severe OI were more often affected by diabetes mellitus (p-value < 0.001), and smoking was more common in Group B compared to the group with lower severity infections (35.2% vs. 16.7%, p-value = 0.028).
總共有 544 名經臨床和放射學診斷為牙源性感染的患者，於 2017 年 1 月至 2022 年 4 月期間在羅馬尼亞 SCMUT 的頜面外科部門入院和住院。只有 108 名患者符合納入標準並納入研究，如圖 1 所示。根據 SS 評分，患者進一步分為兩組，如下所示：A 組——低嚴重程度感染組，包含 54 名患者，其嚴重程度評分在 SS 量表上為 0 到 8 分；B 組——高嚴重程度感染組，包含 54 名患者，其嚴重程度評分為 9 分到超過 16 分。表 2 描述了牙源性感染患者背景特徵的比較。觀察到男性更常發生 OI（A 組為 55.6%，B 組為 66.7%）。A 組的平均年齡為 46.7 歲（年齡範圍 18-81 歲），而 B 組為 51.7 歲（年齡範圍 20-85 歲），無統計學顯著差異（p 值 = 0.150）。然而，研究組之間的原產地存在顯著差異，來自農村地區的患者更常患有更嚴重的感染（68.5% vs. 46.3%，p 值 = 0.019）。此外，患有嚴重 OI 的患者更常受到糖尿病的影響（p 值 < 0.001），並且與嚴重程度較低的感染組相比，B 組的吸煙更為常見（35.2% vs. 16.7%，p 值 = 0.028）。

3.2. Characteristics of Infection in the Study Population
3.2. 研究人群中感染的特征

Regarding the infection type in OI admitted to the hospital, 70.4% of them were abscesses in the lower infection cohort (Group A), while in Group B, 55.6% of infections were associations of abscesses and cellulitis (p-value < 0.001), as seen in Table 3. The most involved infection sites were the superficial lodges (40.7% vs. 48.1%), and peri-mandibular infections (25.9% vs. 33.3%), without statistically significant differences. Regarding disease outcomes, a total of 22.2% of patients in Group B developed sepsis, compared to 7.4% in Group A (p-value = 0.030), and four patients with severe OI were admitted to the ICU. However, mortality was not significantly different between the study groups (0.0% in Group A vs. 5.6% in Group B, p-value = 0.078). The median duration of hospitalization was significantly longer in patients from Group B, compared to Group A (12.0 days vs. 4.1 days, p-value < 0.001), in correlation with a higher frequency of severe complications in Group B (16.7% vs. 3.7%, p-value = 0.025).
至于因牙源性感染 (OI) 入院的感染类型，在较低感染队列（A 组）中，70.4% 为脓肿，而在 B 组中，55.6% 的感染是脓肿和蜂窝组织炎的结合（p 值 < 0.001），如表 3 所示。最常涉及的感染部位是浅表间隙（40.7% vs. 48.1%）和下颌周围感染（25.9% vs. 33.3%），没有统计学上的显着差异。关于疾病结果，B 组共有 22.2% 的患者发展为败血症，而 A 组为 7.4%（p 值 = 0.030），并且有 4 名患有严重 OI 的患者被送入 ICU。然而，研究组之间的死亡率没有显着差异（A 组为 0.0%，B 组为 5.6%，p 值 = 0.078）。与 A 组相比，B 组患者的平均住院时间明显更长（12.0 天 vs. 4.1 天，p 值 < 0.001），这与 B 组严重并发症发生率较高有关（16.7% vs. 3.7%，p 值 = 0.025）。

The symptom severity evaluation presented in Table 4 identified a total of 32 (59.2%) patients with a SIRS score ranging from 0 to 1 in Group A. On the other side, Group B patients were only 13 (24.0%) within the 0–1 score range (p-value < 0.001). A severe trismus score was observed in 27 (50.0%) of patients from Group B, compared to only 9.3% in Group A (p-value < 0.001). Similar observations were noticed in the dysphagia score and fascial space score, where a statistically significantly higher prevalence of high severity was found in Group B patients. The prevalence of patients with odontogenic infections who were admitted with dehydration and significant comorbidities was significantly higher in Group B (29.6% vs. 5.6% in Group A, p-value = 0.001).
表 4 中呈現的症狀嚴重程度評估顯示，A 組共有 32 名患者 (59.2%) 的 SIRS 評分範圍為 0 到 1。另一方面，B 組患者中只有 13 名 (24.0%) 的評分範圍在 0-1 之間（p 值 < 0.001）。 在 B 組患者中觀察到嚴重的牙關緊閉評分為 27 (50.0%)，而 A 組僅為 9.3%（p 值 < 0.001）。 在吞嚥困難評分和筋膜間隙評分中也觀察到類似的現象，B 組患者中高嚴重程度的患病率在統計學上顯著較高。 B 組患者因牙源性感染入院時伴有脫水和顯著合併症的患者患病率顯著較高（B 組為 29.6%，A 組為 5.6%，p 值 = 0.001）。

3.3. Risk Assessment in the Study Population
3.3. 研究人群的風險評估

Table 5 presents the comparison of severity scores and biomarker scores among patients with odontogenic infections admitted to the hospital. It was observed that SS and SII scores were statistically significantly higher among patients in Group B (13.6 vs. 6.1, p-value < 0.001), respectively, 2312.4 in Group B compared to 696.3 in Group A (p-value < 0.001). All tested biomarker scores were significantly higher in Group B patients, including the CRP-NLR association, with a median score of 341.4, compared with 79.0 in Group B (p-value < 0.001).
表 5 顯示了因牙源性感染入院的患者的嚴重程度評分和生物標誌物評分的比較。觀察到 B 組患者的 SS 和 SII 評分在統計學上顯著高於 A 組（分別為 13.6 vs. 6.1，p 值 < 0.001），B 組為 2312.4，而 A 組為 696.3（p 值 < 0.001）。所有測試的生物標誌物評分在 B 組患者中均顯著較高，包括 CRP-NLR 關聯，中位數評分為 341.4，而 B 組為 79.0（p 值 < 0.001）。

The logistic regression analysis presented in Table 6 describes the predictive of biological markers on the severity of odontogenic infections represented on the SS scale. It was observed that patients with an elevated WBC count had a 5.54 higher likelihood of severe OI, elevated neutrophils (OR = 7.10), elevated lymphocyte count (OR = 8.62), elevated NLR with an odds ratio of 4.46 (p-value < 0.001), high CRP levels with a 6.65 higher likelihood of severe OI, and lastly, the CRP-NLR association being responsible for a 7.28 higher risk (95% CI = 4.83–10.16). The ROC analysis of CRP-NLR resulted in a 0.889 AUC value (p-value < 0.001), with high sensitivity (79.6%) and high specificity (85.1%) for predicting a severe odontogenic infection using these biomarkers measured at hospital admission (Figure 2).
表 6 所示的邏輯回歸分析描述了生物標記物對 SS 量表上表示的齒源性感染嚴重程度的預測能力。觀察到白血球計數升高的患者患嚴重 OI 的可能性高 5.54 倍，嗜中性球升高（OR = 7.10），淋巴球計數升高（OR = 8.62），NLR 升高，優勢比為 4.46（p 值<0.001），高 CRP 水平患嚴重 OI 的可能性高 6.65 倍，最後，CRP-NLR 的關聯導致風險高 7.28 倍（95% CI = 4.83–10.16）。CRP-NLR 的 ROC 分析得出 0.889 的 AUC 值（p 值<0.001），具有高靈敏度（79.6%）和高特異性（85.1%），可用於預測入院時測量的這些生物標記物所導致的嚴重齒源性感染（圖 2）。

4.1. Important Findings  4.1. 重要發現

The decision-making process in medicine incorporates clinical and laboratory considerations. Detecting an increase in acute phase reactants may assist the diagnostic interpretation of clinical symptoms in circumstances when an infection is suspected. In our investigation, CRP-NLR, which consists of CRP level at admission and NLR level at admission, was shown to have a more accurate ability to predict the severity of odontogenic infection. Initially, it was shown that a high CRP-NLR was substantially and strongly connected with high severity levels in odontogenic infection. Then, we examined the connection between CRP-NLR levels and WBC levels at admission in both severity groups and discovered that CRP-NLR had a greater predictive capability.
醫學上的決策過程包含臨床和實驗室的考量。當懷疑感染時，檢測急性期反應物的增加可能有助於臨床症狀的診斷解釋。在我們的研究中，CRP-NLR（包括入院時的 CRP 水平和入院時的 NLR 水平）顯示出更準確的預測牙源性感染嚴重程度的能力。最初，研究表明，高 CRP-NLR 與牙源性感染的高嚴重程度水平顯著且強烈相關。然後，我們檢查了 CRP-NLR 水平與兩個嚴重程度組的入院時的 WBC 水平之間的聯繫，並發現 CRP-NLR 具有更高的預測能力。

A better understanding of the inflammatory cascade has led to new discoveries and the identification of many mediators that, in combination with clinical symptoms, might serve as valuable infection indicators [36]. Bagul et al. [37] concluded in their study that CRP should be recommended as a monitoring marker for managing patients with fascial space infections of odontogenic origin, as it is a more sensitive indicator than WBC count and one of the best measuring tools for determining the infection control in these patients. In addition, John CR et al. [38] showed in their research analyzing indicators in patients with odontogenic fascial space infections that CRP should be suggested as a monitoring marker for the diagnosis of fascial space infection and for determining the response to treatment. In their research, Barreto et al. [39] found that the CRP test is a practical, easily accessible blood test that portrays the patient course and response to therapy more precisely than other commonly used indicators in oral and maxillofacial surgery.
對於炎症級聯反應的更深入理解，促成了新的發現，並確定了許多與臨床症狀相結合可能成為有價值的感染指標的介質[36]。Bagul 等人[37]在他們的研究中得出結論，建議將 CRP 作為監測標記，用於管理患有牙源性筋膜間隙感染的患者，因為它比 WBC 計數更敏感，並且是確定這些患者感染控制的最佳測量工具之一。此外，John CR 等人[38]在他們的研究中分析了牙源性筋膜間隙感染患者的指標，表明應將 CRP 作為監測標記，用於診斷筋膜間隙感染和確定對治療的反應。Barreto 等人[39]在他們的研究中發現，CRP 測試是一種實用且易於獲得的血液測試，與口腔頜面外科中其他常用的指標相比，它更準確地描述了患者的病程和對治療的反應。

Dynamic changes in NLR, on the other hand, predate the clinical condition by several hours and may alert doctors to an ongoing pathogenic process. Despite these benefits, NLR as a biomarker for assessing the progression of odontogenic infection has limited use. A recent meta-analysis [40] revealed that NLR was greater in non-survivors of sepsis than in survivors, and a larger NLR was linked with a worse prognosis in sepsis patients. Independent of the kind of operation (cardiac or abdominal), preoperative NLR levels are independent predictors of postoperative problems [41,42,43].
另一方面，NLR 的動態變化比臨床狀況早幾個小時出現，可能會提醒醫生注意正在進行的致病過程。儘管有這些好處，但 NLR 作為評估牙源性感染進展的生物標誌物的用途有限。最近的一項薈萃分析 [40] 顯示，非敗血症存活者的 NLR 高於敗血症存活者，而較高的 NLR 與敗血症患者的較差預後有關。與手術類型（心臟或腹部）無關，術前 NLR 水平是術後問題的獨立預測因子 [41, 42, 43]。

In addition, NLR may be used as a predictor for surgical treatment in submandibular abscesses [44] and as a recovery marker in odontogenic infection. Several investigations have demonstrated a correlation between NLR and the occurrence of pus, duration of hospital stay, and antibiotic dosage need [45]. In addition, the NLR value is constant and resistant to physiological and environmental factors, such as dehydration, physical exercise, and blood sample processing, that might influence test findings [46]. In their investigation, Dogruel et al. [47] determined that the NLR was related to hospitalization and antibiotic dosages in individuals with odontogenic infection. Incorporating the NLR into the CRP level has tremendous promise as a biomarker for odontogenic infection severity classification. Our objective was to determine whether CRP and NLR may serve as possible severity indicators in patients with odontogenic infections (OI).
此外，NLR 可用作頜下腺膿腫手術治療的預測因子 [ 44]，以及牙源性感染的恢復指標。多項研究表明，NLR 與膿液的產生、住院時間和抗生素劑量需求之間存在相關性 [ 45]。此外，NLR 值穩定，且不受生理和環境因素的影響，如脫水、體力鍛鍊和血液樣本處理等，這些因素可能會影響測試結果 [ 46]。Dogruel 等人 [ 47] 在他們的研究中確定，NLR 與牙源性感染患者的住院治療和抗生素劑量有關。將 NLR 納入 CRP 水平中，作為牙源性感染嚴重程度分類的生物標誌物具有巨大的前景。我們的目標是確定 CRP 和 NLR 是否可以作為牙源性感染（OI）患者的潛在嚴重程度指標。

According to several research, despite the increase in frequency, the patient features have remained basically unchanged [48]. The majority of patients were in their mid-30s, which is much younger than the majority of patients in our research, who were in their mid-40s. Furthermore, the amount of time between the beginning of symptoms and hospital presentation stayed comparable in both groups, and the percentage of patients who sought dental treatment prior to hospitalization remained surprisingly high at more than 40%, despite the fact that our study lacks this type of information. In addition, almost two-thirds of patients reported in previous studies had been orally administered antibiotics by their dentist or primary care physician before presenting to the hospital, a number that increased from 57% to 63%. Instead of seeking to cure the underlying cause, it is considered that an over dependence on antibiotics results in suboptimal patient care. Therefore, the need for prediction scores and algorithms is essential to determine the patients at risk.
根據多項研究，儘管發生頻率有所增加，但患者的特徵基本上保持不變 [ 48]。大多數患者年齡在 30 多歲，遠低於我們研究中的大多數患者，他們的年齡在 40 多歲。此外，從症狀開始到入院的時間長度在兩組中保持相似，並且在入院前尋求牙科治療的患者比例仍然高得驚人，超過 40％，儘管我們的研究缺乏這類信息。此外，先前研究中報告的幾乎三分之二的患者在入院前已經由他們的牙醫或初級保健醫生口服給予抗生素，這個數字從 57％增加到 63％。人們認為，過度依賴抗生素而不是尋求治療根本原因，會導致次優的患者護理。因此，需要預測評分和算法來確定有風險的患者。

4.2. Limitations of the Study
4. 2. 研究的局限性

Our research has some important limitations and restrictions. To begin, the study was a single-center investigation of patients who had been admitted to the medical facility for odontogenic infections. Second, because of the retrospective design, we had to rely on the data from medical records; as a result, statistical analysis was susceptible to the risk of being inaccurate due to human error. Additionally, the retrospective study design impacts our results, as the research depends on the accuracy of both patient information tracking and the digital transcription of data from paper records. Limited by the retrospective design of our study, we could not perform a dynamic profile analysis of CRP and NLR, which may offer more helpful information. Other limitations are represented by country-specific features, since all patients were from Romania, and the oral hygiene can influence the severity of odontogenic infections. To provide more evidence in support of our results, further prospective research should be carried out.
我們的研究有一些重要的局限性和限制。首先，該研究是對因牙源性感染而入住醫療機構的患者進行的單中心調查。其次，由於採用回顧性設計，我們不得不依賴醫療記錄中的數據；因此，統計分析容易因人為錯誤而產生不準確的風險。此外，回顧性研究設計會影響我們的結果，因為研究依賴於患者信息跟踪和從紙質記錄中數位轉錄數據的準確性。由於我們研究的回顧性設計的限制，我們無法對 CRP 和 NLR 進行動態概況分析，這可能會提供更有幫助的信息。其他限制因素是國家/地區的特定特徵，因為所有患者都來自羅馬尼亞，而且口腔衛生會影響牙源性感染的嚴重程度。為了提供更多證據來支持我們的結果，應進行進一步的前瞻性研究。