Introduction
Acute tonsillitis is frequently encountered in otolaryngology departments in hospitals worldwide. Some patients exhibit disease aggravation, which culminates in the formation of a peritonsillar abscess.Reference Raut1 An accurate diagnosis of peritonsillar abscess in patients with tonsillitis is often difficult, because peritonsillar swellings are not always attributed to an abscess.Reference Mohamad and Yaroko2
Mean platelet volume is a parameter evaluated in complete blood count studies, and it represents variances in the platelet size.Reference Ozdemir, Karadeniz, Doksoz, Celegen, Yozgat and Guven3 Of late, the significance of mean platelet volume in inflammatory conditions has been widely investigated.Reference Gasparyan, Ayvazyan, Mikhailidis and Kitas4 Inflammation promotes platelet activation. An increase in mean platelet volume is a marker of platelet activation in the initial phase of infection. In inflammation, a decrease in mean platelet volume is observed after the utilisation of activated or large platelets, either because of the depletion of these platelets and/or defective thrombopoiesis.Reference Eryilmaz and Derin5
Recently, Şentürk et al. reported that mean platelet volume was significantly higher in the peritonsillar abscess group than in the control group, and it showed a significant decrease after treatment.Reference Şentürk, Azgın, Övet, Alataş, Ağırgöl and Yılmaz6 To the best of our knowledge, no other study has investigated the relationship between tonsillitis and mean platelet volume. Therefore, the present study was conducted to determine the diagnostic role of mean platelet volume in tonsillitis with and without peritonsillar abscess.
Materials and methods
Study population
In this case–control study, data for patients who were diagnosed with severe tonsillitis at our hospital between May 2011 and September 2016 were retrospectively reviewed. Severe tonsillitis was defined as marked tonsillar erythema and/or swelling that caused a disruption in daily life and activities.Reference Suzumoto, Hotomi, Billal, Fujihara, Harabuchi and Yamanaka7 Peritonsillar abscess was diagnosed using computed tomography (CT), the findings of which were confirmed by aspiration or incision and drainage.
Data regarding age, sex and clinical history were collected. Data for patients who did not undergo CT, those with mild to moderate tonsillitis that did not require blood tests, and those with infectious mononucleosis were excluded from analyses.
Healthy individuals who underwent routine medical check-ups at our hospital between April 2014 and May 2014 were enrolled as controls.
The study protocol was approved by the ethics review committee of our hospital (approval number: 2015–1), and was conducted in accordance with the code of ethics of the World Medical Association (Declaration of Helsinki). The requirement for informed consent was waived because of the retrospective study design.
Biomarker analysis
Venous blood samples were obtained at the time of physical examination. Complete blood count studies were performed using the UniCel DxH 800 haematology analyser (Beckman Coulter, Brea, California, USA). The reference range for a normal mean platelet volume with the UniCel DxH 800 is 7.9–10.6 fl. C-reactive protein (CRP) was measured using the TBA-c16000 automated clinical chemistry analyser (Toshiba Medical Systems, Tochigi, Japan). The CRP was not examined for the control group because it indicates acute inflammation. The neutrophil–lymphocyte ratio, a potential marker of inflammation that is increased in patients with tonsillitis, was calculated from the neutrophil and lymphocyte counts.
Statistical analysis
All statistical analyses were performed using SAS 9.4 software (SAS Institute, Cary, North Carolina, USA). Continuous variables are presented as means ± standard deviations, and categorical variables are expressed as frequencies (percentages). The diagnostic accuracy of the biomarkers was evaluated using the Mann–Whitney U test or a chi-square test. Receiver operating characteristic curve analysis was used to determine the association of mean platelet volume with tonsillitis. A p-value of less than 0.05 was considered statistically significant.
Results
Study patients
During the study period, 31 patients were diagnosed with tonsillitis, 1 of whom did not undergo CT and was excluded. In the 30 included patients, 14 were diagnosed with peritonsillar abscess. We further divided the patients into those with peritonsillar abscess (n = 14) and those without (n = 16).
Mean platelet volume and tonsillitis
The mean platelet volume was significantly lower in the tonsillitis group (7.8 ± 0.7 per cent) than in the control group (8.7 ± 0.6 per cent; p < 0.0001; Table I), while white blood cell (WBC) counts and neutrophil–lymphocyte ratio were significantly higher in the former group than in the latter (Table I).
Demographic and haematological parameters for tonsillitis patients and healthy controls
Data are expressed as means ± standard deviations, unless indicated otherwise. *n = 30; †n = 62. RBC = red blood cell count; Hb = serum haemoglobin; WBC = white blood cell; NLR = neutrophil to lymphocyte ratio; MPV = mean platelet volume
The demographic and clinical characteristics of the patients with tonsillitis are summarised in Table II. There were no differences in clinical characteristics between patients with peritonsillar abscess and those without (Table II).
Demographic and clinical characteristics of tonsillitis patients with and without peritonsillar abscess
*n = 14; †n = 16. SD = standard deviation
Mean platelet volume was significantly lower in patients with peritonsillar abscess (7.5 ± 0.6 per cent) than in those without (8.0 ± 0.7 per cent; p = 0.0277; Table III). There were no significant differences in any other laboratory parameter between the two groups (Table III), which showed similar WBC counts. The CRP was higher in patients with peritonsillar abscess than in those without, although the difference was not statistically significant.
Haematological parameters for tonsillitis patients with and without peritonsillar abscess
Data are expressed as means ± standard deviations, unless indicated otherwise. *n = 14; †n = 16. RBC = red blood cell count; Hb = serum haemoglobin; WBC = white blood cell; NLR = neutrophil to lymphocyte ratio; MPV = mean platelet volume; CRP = C-reactive protein
Predictive value of mean platelet volume
Receiver operating characteristic curve analysis revealed a mean platelet volume cut-off value of 7.95 fl for the diagnosis of tonsillitis, with a sensitivity and specificity of 67 per cent and 92 per cent, respectively (area under the curve, 0.8323; Figure 1). The cut-off value for the diagnosis of peritonsillar abscess was 7.75 fl, with a sensitivity and specificity of 86 per cent and 69 per cent, respectively (area under the curve, 0.7478; Figure 2).
Receiver operating characteristic curve, used to determine the mean platelet volume cut-off value for the diagnosis of tonsillitis. The mean platelet volume cut-off value was 7.95 fl, with a sensitivity and specificity of 67 per cent and 92 per cent respectively (area under curve, 0.8323; 95 per cent confidence interval = 0.7345–0.9311).
Receiver operating characteristic curve, used to determine the mean platelet volume cut-off value for the diagnosis of abscess associated with tonsillitis. The mean platelet volume cut-off value was 7.75 fl, with a sensitivity and specificity of 86 per cent and 69 per cent respectively (area under curve, 0.7478; 95 per cent confidence interval = 0.5604–0.9352).
Discussion
In the present study, we assessed the diagnostic role of mean platelet volume in tonsillitis patients with and without peritonsillar abscess. The results revealed that mean platelet volume was significantly lower in patients with tonsillitis than in those without. Moreover, it was significantly lower in patients with peritonsillar abscess than in those without. However, WBC counts were similar in patients with and without peritonsillar abscess. The CRP was higher in the peritonsillar abscess group, although the difference was not statistically significant. These results suggest that the diagnostic accuracy of mean platelet volume for tonsillitis and peritonsillar abscess is higher than that of WBC counts and CRP levels. Thus, mean platelet volume can be a useful diagnostic marker of tonsillitis with and without peritonsillar abscess.
Although the mechanism underlying the decrease in mean platelet volume in patients with tonsillitis and peritonsillar abscess remains unclear, severe inflammation seems to be the most likely cause. Mean platelet volume has been reported as a novel inflammatory biomarker.Reference Arslan, Çandar, Kuran, Akmansu and Kocatürk8, Reference Mert and Terzi9 While mean platelet volume has been reported to increase in some inflammatory conditions such as obstructive sleep apnoea syndrome and idiopathic sudden hearing loss,Reference Günbey, Karabulut, Karabulut and Zaim10, Reference Ulu, Ulu, Ahsen, Yucedag, Aycicek and Celik11 it has been reported to decrease in acute appendicitis.Reference Kucuk and Kucuk12 Appendicitis is an inflammatory process affecting the lymphoid tissues, and the pathological changes are similar to those occurring in tonsillitis.Reference Hunt13 Other inflammatory conditions with a low mean platelet volume include exacerbated asthma and acute rotavirus gastroenteritis.Reference Sun, Zhang, Cao, Li and Wang14, Reference Mete, Akelma, Cizmeci, Bozkaya and Kanburoglu15
The reasons why mean platelet volume is high in certain diseases and low in others remain unclear. Gurler and Aktas speculated that inflammatory cytokines in the bone marrow affect the increase or decrease in mean platelet volume.Reference Gurler and Aktas16 In patients with chronic tonsillitis, interleukin (IL)-1b and IL-6 levels were found to be significantly higher than those in control patients.Reference Unal, Oztürk and Görür17 Activated platelets increase in size in the initial phase of inflammation, leading to an increase in mean platelet volume. Subsequently, at an uncertain cytokine level, mean platelet volume decreases because of the utilisation and depletion of these activated, large platelets and/or defective thrombopoiesis; the remaining small platelets lower mean platelet volume.Reference Gurler and Aktas16
The findings in our study are inconsistent with those reported by Şentürk et al.Reference Şentürk, Azgın, Övet, Alataş, Ağırgöl and Yılmaz6 The reasons for this discrepancy are not clearly understood. One possible explanation is the difference in the timing of blood collection. In the present study, blood was collected at an average of 4.2 days after disease onset; the timing of blood collection in the previous study is unknown. Another explanation is the effects of age and sex on mean platelet volume, although no conclusive data are available.Reference Noris, Melazzini and Balduini18 Patients with peritonsillar abscess in the present study were older than those in the previous study (46.7 ± 15.5 years vs 34.6 ± 12.8 years), which also included more men compared with the present study (12 men and 2 women vs 46 men and 42 women; 85.7 per cent men vs 52.3 per cent men).
A thorough clinical examination remains the ‘gold standard’ for the diagnosis of peritonsillar abscess. Although the performance of CT may not be usual practice,Reference Steyer19, Reference Powell and Wilson20 it is frequently implemented in several countries. This is because CT is able to: (1) differentiate cellulitis from abscess; (2) define the precise location and extent of complex infections; (3) identify any secondary complications; and (4) occasionally suggest the infection source.Reference Patel, Ahmad, O'Leary and Michel21–Reference Teschner, Aljeraisi, Giesemann, Götz, Lenarz and Kontorinis24 Accordingly, we performed CT for the evaluation of peritonsillar abscess in the present study.
Our study has several limitations. First, data were retrospectively collected. Second, the sample size was small. Third, the ratio of men to women was not balanced. Fourth, we did not evaluate follow-up data after treatment for tonsillitis. Therefore, further prospective studies with post-treatment data and larger sample sizes are necessary to validate our findings.
• Tonsillitis is frequently encountered in ENT departments worldwide
• The diagnostic role of mean platelet volume was investigated in tonsillitis patients with and without peritonsillar abscess
• Mean platelet volume was decreased in tonsillitis patients, and was lower in those with peritonsillar abscess than in those without
• Mean platelet volume measurement can provide useful diagnostic information for physicians treating tonsillitis patients
Conclusion
In conclusion, to the best of our knowledge, our study is the first to show that mean platelet volume is significantly decreased in patients with tonsillitis, being significantly lower in patients with peritonsillar abscess than in those without. Therefore, measurement of mean platelet volume can provide useful diagnostic information for physicians treating patients with tonsillitis. We are currently planning to conduct a larger multicentre study to explore the role of other biomarkers in addition to mean platelet volume.
Competing interests
None declared.
