**2. Predictors for PHN**

Previous studies have shown that older age, female sex, presence of a prodrome, greater rash severity, and greater acute pain severity are predictors of increased PHN [14-18, 25]. Some other potential predictors (ophthalmic localization, presence of anxiety and depression, presence of allodynia, and serological/virological factors) have also been studied [14, 18]. We conducted a retrospective study to identify significant predictors of PHN that would

contribute to the establishment of evidence-based medicine approaches to the optimal treatment of PHN [24].

(0/1) or (0/1/2), median (range)

http://dx.doi.org/10.5772/55278

149

Predictive Factors for Postherpetic Neuralgia and Recent Pharmacotherapies

**CI of OR Lower 95% Upper 95%**

*Period of onset, type and extent of pain*

VAS, visual analog scale

\*P<0.05

addition of prodromal pain and allodynia [24].

Period before herpes zoster onset (days) 30 (1-3000) Log (period before herpes zoster onset) 1.477 (0-3.477)

Binary scales were: female = 0, male = 1 for sex; and absent = 0, present = 1 for others.

**Variable EV SE χ2 value P OR**

EV, estimated value; SE, standard error; OR, odds ratio; CI, confidence interval; DM, diabetes mellitus

**Table 2.** Results of multivariate ordered logistic regression analysis for variables extracted by forward selection with

associated pain; and <50 years = 0, 51-74 years = 1, ≥75 years = 2 for age.

The ordered scale was: absent after 3 months = 0, present = 1, and present after 6 months = 2 for significant zoster-

**Table 1.** Clinical characteristics of 73 patients and various factors that could be related to the occurrence of PHN [24]

*Age 1.008 0.461 4.78 0.0288\* 2.740 1.110 6.761* Prodromal pain 0.442 0.532 0.69 0.4059 1.556 0.549 4.413 DM 1.123 0.693 2.63 0.1049 3.075 0.791 11.952 Allodynia 0.201 0.630 0.10 0.7489 0.818 0.238 2.808 Pain reduced by bathing 1.221 0.745 2.68 0.1014 3.389 0.787 14.601 *Deep pain 1.446 0.682 4.49 0.0341\* 4.244 1.114 16.163* Breakthrough pain 0.687 0.598 1.32 0.2506 1.988 0.616 6.418 Sleep disorder 0.146 0.514 0.08 0.7757 1.158 0.423 3.169

Prodromal rash 39/34 Prodromal pain 34/39 Allodynia 35/38 VAS (mm) 66 (0-100) Pain reduced when bathing 41/32 Pain (superficial) 35/38 Pain (deep) 38/35 Pain (continuous) 43/30 Pain (breakthrough) 31/42

The participants were 73 patients with herpes zoster who had been treated at the pain clinic of our hospital between January 2008 and June 2010. Variables present at the initial visit were extracted from the clinical records for regression analysis of factors related to the occurrence of PHN. The following scores for response were used: 0 = no PHN after 3 months; 1 = PHN present after 3 months, but absent after 6 months; and 2 = PHN present after 6 months. Multivariate ordered logistic regression analysis was performed to identify predictive factors for PHN. Table 1 shows the clinical characteristics of patients and various factors that could be related to the occurrence of PHN. Multivariate ordered logistic regression analysis identi‐ fied advanced age and deep pain at the initial visit to our outpatient pain clinic as factors predicting the occurrence of PHN. diabetes mellitus (DM) and pain reduced by bathing also showed high odds ratios, but were not significant predictive factors (Table 2). In conclusion, advanced age and deep pain at first visit were identified as predictive factors for PHN. DM and pain reduced by bathing should also be considered as potential predictors of PHN [24]. To improve patient safety and the likelihood of achieving satisfactory outcomes, prospective studies are needed to establish optimal treatments including pharmacotherapy for PHN. Next section we further review recent pharmacotherapeutic results for PHN.



Binary scales were: female = 0, male = 1 for sex; and absent = 0, present = 1 for others.

The ordered scale was: absent after 3 months = 0, present = 1, and present after 6 months = 2 for significant zosterassociated pain; and <50 years = 0, 51-74 years = 1, ≥75 years = 2 for age.

VAS, visual analog scale

contribute to the establishment of evidence-based medicine approaches to the optimal

148 Peripheral Neuropathy - A New Insight into the Mechanism, Evaluation and Management of a Complex Disorder

The participants were 73 patients with herpes zoster who had been treated at the pain clinic of our hospital between January 2008 and June 2010. Variables present at the initial visit were extracted from the clinical records for regression analysis of factors related to the occurrence of PHN. The following scores for response were used: 0 = no PHN after 3 months; 1 = PHN present after 3 months, but absent after 6 months; and 2 = PHN present after 6 months. Multivariate ordered logistic regression analysis was performed to identify predictive factors for PHN. Table 1 shows the clinical characteristics of patients and various factors that could be related to the occurrence of PHN. Multivariate ordered logistic regression analysis identi‐ fied advanced age and deep pain at the initial visit to our outpatient pain clinic as factors predicting the occurrence of PHN. diabetes mellitus (DM) and pain reduced by bathing also showed high odds ratios, but were not significant predictive factors (Table 2). In conclusion, advanced age and deep pain at first visit were identified as predictive factors for PHN. DM and pain reduced by bathing should also be considered as potential predictors of PHN [24]. To improve patient safety and the likelihood of achieving satisfactory outcomes, prospective studies are needed to establish optimal treatments including pharmacotherapy for PHN. Next

(0/1) or (0/1/2), median (range)

section we further review recent pharmacotherapeutic results for PHN.

Age 8/46/19, 69 (27-90)

Significant zoster-associated pain 35/13/25

Sex 37/36

Hypertension 44/29 Angina 69/4 Diabetes mellitus 61/12 Malignant tumor 50/23 Autoimmune disease 68/5 Sleep disorder 36/37

Trigeminal nerves 58/15 Cervical nerves 58/15 Thoracic nerves 42/31 Lumbar nerves 61/12 Sacral nerves 70/3

treatment of PHN [24].

*Demographics*

*Complication*

*Location*

**Table 1.** Clinical characteristics of 73 patients and various factors that could be related to the occurrence of PHN [24]


EV, estimated value; SE, standard error; OR, odds ratio; CI, confidence interval; DM, diabetes mellitus

\*P<0.05

**Table 2.** Results of multivariate ordered logistic regression analysis for variables extracted by forward selection with addition of prodromal pain and allodynia [24].
