Predicting Postherpetic Neuralgia in Patients with Herpes Zoster by Machine Learning: A Retrospective Study

Wang, Xinxing; Zhang, Yi; Fan, Bifa

doi:10.1007/s40122-020-00196-y

Cited by 24 publications

(23 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The literature exclusion criteria were as follows: (I) data had significant errors, errors, or low reliability; (II) improper use of literature research statistics; (III) documents were repeatedly published or only used for academic meetings and exchanges; (IV) the main content of the literature deviated from the research direction of this paper; (V) the literature did not contain the relevant elements of this study. A total of 14 studies were included in this meta-analysis (1,5,(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18).…”

Section: Literature Screeningmentioning

confidence: 99%

See 1 more Smart Citation

A systematic review and meta-analysis of independent risk factors for postherpetic neuralgia

Zhou¹,

Jin²,

Chen³

et al. 2021

Ann Palliat Med

View full text Add to dashboard Cite

Background: This study aimed to explore the independent risk factors for postherpetic neuralgia (PHN).Methods: Related studies of PHN risk factors were searched in PubMed for screening and meta-analysis.In this study, data from studies included were extracted and summarized, including odds ratio (OR) value, 95% confidence interval (CI), P value, sample size, and the number of patients with and without PHN. The chi-square test was used for heterogeneity test. Sensitivity analysis was conducted by excluding low-quality studies and using different model analysis.Results: A total of 14 studies were further screened for meta-analysis, including 4,192 patients with herpes zoster. Of these patients, 478 (11.40%) had neuralgia and 3,714 (88.60%) did not have neuralgia. Age [OR =1.59; 95% CI: (1.23, 2.04); Z=3.62; P<0.001], acute severe pain in the herpes stage [OR =1.49; 95% CI: (1.08, 2.08); Z=2.39; P=0.02], prodromal symptoms [OR =2.00; 95% CI: (1.16, 3.44); Z=2.48; P=0.01], and severe rash [OR =2.40; 95% CI: (1.83, 3.14); Z=6.38; P<0.001] were independent risk factors for PHN. The funnel chart shows that there is no publication bias or geographic bias in the above independent risk factors. Gender (Z=0.37; P=0.71) was not associated with PHN, and the funnel chart shows that there is no publication bias or geographic bias.Discussion: Age, acute pain, prodromal symptoms, and severe rash were independent risk factors for PHN.

show abstract

Section: Literature Screeningmentioning

confidence: 99%

“…Postherpetic neuralgia (PHN) is a common sequela of herpes zoster (1,2). PHN lacks a unified definition, mainly referring to the recurrence of pain symptoms after the local rash subsides or skin lesions heal.…”

Section: Introductionmentioning

confidence: 99%

A systematic review and meta-analysis of independent risk factors for postherpetic neuralgia

Zhou¹,

Jin²,

Chen³

et al. 2021

Ann Palliat Med

View full text Add to dashboard Cite

show abstract

“…For patients with herpes zoster, the area of rash is an independent factor affecting the occurrence of postherpetic neuralgia. [31][32][33] The smaller the lesion area is, the less likely it is to develop postherpetic neuralgia. For patients with limb herpes zoster, the number of affected centrums ≤ 3 is also a protective factor for motor dysfunction in patients with limb herpes zoster.…”

Section: Discussionmentioning

confidence: 99%

Analysis of Risk Factors and Construction of a Prediction Model of Motor Dysfunction Caused by Limb Herpes Zoster

Tang¹,

Tao²,

Luo³

et al. 2022

JPR

View full text Add to dashboard Cite

To analyze the influencing factors of motor dysfunction in patients with limb herpes zoster and to construct a nomogram to predict the risk of this complication. Methods: A total of 213 patients with limb herpes zoster in the First Hospital of Jiaxing from January 2015 to June 2020 were enrolled. The muscle strength of the affected limb was evaluated by the Medical Research Council (MRC) score. The muscle strength of the affected limb was less than grade 5, and lower than that of the contralateral limb (normal side) was defined as motor dysfunction. Univariate and multivariate logistic regression analyses were used to determine the influencing factors of motor dysfunction in patients with limb herpes zoster. The nomogram of the risk of motor dysfunction was drawn by R software. Discrimination and calibration of the prediction model were assessed by using the C-index and calibration plot. Internal validation was assessed by using bootstrapping validation. Results: There were 58 cases in the motor dysfunction group and 155 cases in the nonmotor dysfunction group, and univariate analysis showed that the location of involvement (OR = 4.095, 95% CI: 2.097-8.466, P = 0.017), whether or not diseases were combined (OR = 0.520, 95% CI: 0.281-0.959, P = 0.036) and the number of affected centrums (OR = 0.336, 95% CI: 0.177-0.632, P = 0.001) were associated with the occurrence of motor dysfunction in patients with limb herpes zoster. Multivariate logistic regression analysis showed that upper limb involvement (OR = 3.811, 95% CI: 1.829-8.387, P = 0.001), noncombined diseases (OR = 0.493, 95% CI: 0.249-0.969, P = 0.041) and the number of affected centrums ≤3 (OR = 0.439, 95% CI: 0.218-0.881, P = 0.020) were independent influencing factors of this complication. The prediction model displayed good discrimination with a C-index of 0.72 and good calibration. A high C-index value of 0.71 could still be reached in the interval validation. Conclusion:This study used the clinical data available to establish a nomogram model for motor dysfunction of limb herpes zoster and found that motor dysfunction is more likely to occur in patient who meets upper limb involvement, combined diseases and the number of affected centrums >3. This kind of high-risk group should be intervened as soon as possible.

show abstract

“…For ease of calculation, we conducted the following data preprocessing: In terms of sex, we marked “1” for female and “2” for male. The guide methods of the procedure include ultrasound recorded as “1,” CT + ultrasound recorded as “2,” radiography + ultrasound recorded as “3,” CT recorded as “4,” and radiography recorded as “5.” As for education level, we recorded illiteracy as “1,” elementary school level as “2,” junior high school as “3,” senior high school as “4,” and university and above as “5.” We divided different stages of HZ into three layers, which were represented by consecutive numbers: ≤ 1 month was recorded as “1,” between 1 and 3 months was recorded as “2,” and > 3 months was recorded as “3.” As for rash site, we recorded the cervical segment as “1,” head segments as “2,” lumbosacral segments as “3,” and thoracic segment as “4.” For the remaining features, we denoted “no” as “0” and “yes” as “1.” To describe effectiveness of the PRF treatment, we marked the negative as “0” and the positive as “1” according to general practices [ 7 ].…”

Section: Methodsmentioning

confidence: 99%

“…For the remaining features, we denoted ''no'' as ''0'' and ''yes'' as ''1.'' To describe effectiveness of the PRF treatment, we marked the negative as ''0'' and the positive as ''1'' according to general practices [7].…”

Section: Data Processingmentioning

confidence: 99%

Development of a Model for Predicting the Effectiveness of Pulsed Radiofrequency on Zoster-Associated Pain

et al. 2022

View full text Add to dashboard Cite

Introduction: Zoster-associated pain (ZAP), which may cause anxiety, depression, and sleep disorders and reduce quality of life, is often refractory to current standard treatments. Studies have shown that pulsed radiofrequency (PRF) can alleviate ZAP and reduce the incidence of postherpetic neuralgia (PHN). This study aimed to explore the clinical characteristics associated with PRF responsiveness, develop a model for identifying risk factors of inadequate PRF management, and help clinicians make better decisions. Methods: Patients who underwent PRF for ZAP between January 2017 and October 2020 in our hospital were included in this study. Patients were evaluated using the numerical rating scale (NRS), Insomnia Severity Index, Patient Health Questionnaire-9, and 36-Item Short Form Health Survey (SF-36) before and 3 months after the procedure. Patient demographic data and blood test results were also collected. We defined the effectiveness of PRF for ZAP as relief of [ 50% in NRS scores compared to pre-PRF. Least absolute shrinkage and selection operator (LASSO) regression analyses were subsequently performed to identify factors related to the therapeutic effect of PRF in patients with ZAP. The performance of the prediction model was assessed by the area under the receiver operating characteristic curve (AUC). Results: The effectiveness of PRF in patients with ZAP was 69.6% (total 313 patients) after 3 months. LASSO regression analysis extracted the seven most powerful features in the developed prediction model: sex, stage of herpes zoster (HZ), pregabalin dose, bodily pain indicators of SF-36, lymphocyte count, and low-density lipoprotein cholesterol (LDLC) and complement C4 in peripheral blood. Model = 1.586 ? 0.148 9 lymphocyte ? (-0.001) 9 bodily pain indicators of SF-36 ? (-0.001) 9 pregabalin dose ? 0.028 9 LDLC ? 0.001 9 C4 ? (-0.508) 9 sex ? (-0.128) 9 stage of HZ. We generated the ROC curve for the prediction model, and the final AUC was 0.701. The sensitivity, specificity, and overall accuracy of the model were 90%, 33%, and 73%, respectively. Conclusions: Seven factors were significantly associated with poor PRF outcome: male sex,

show abstract

Predicting Postherpetic Neuralgia in Patients with Herpes Zoster by Machine Learning: A Retrospective Study

Cited by 24 publications

References 24 publications

A systematic review and meta-analysis of independent risk factors for postherpetic neuralgia

A systematic review and meta-analysis of independent risk factors for postherpetic neuralgia

Analysis of Risk Factors and Construction of a Prediction Model of Motor Dysfunction Caused by Limb Herpes Zoster

Development of a Model for Predicting the Effectiveness of Pulsed Radiofrequency on Zoster-Associated Pain

Contact Info

Product

Resources

About