Epidural steroid injections are known to be the most effective method for the treatment of the back pain and radicular pain caused by herniated disks or spinal stenosis. The injection methods can be divided into the dorsal epidural approach such as interlamina and ventral epidural approach such as transforaminal injection. In general, it is known that drug injection into the ventral epidural space is more effective. This is possible in thoracic and lumbar region, but it is so dangerous in cervical region for vertebral artery. This author has tried the transforaminal epdidural approach in upper thoracic region as an alternative methods for injecting drugs into cervical ventral space in patient with unsatisfactory to interlamina cervical epidural injection or cervical transforaminal epidural injection. This study is to evaluate the effectiveness of thoracic approach by analysing the outpatient progress note and PACS data.
By using PACS data, I selected 130 patients who visited our clinic from May 1, 2018 to March 31, 2019. Among them, patients diagnosed with herpes zoster, post-herpetic neuralgia were excluded, and in case of multiple injections in the same patient, only the first case was included. If the image quality is not good to determine the success of the injection, excluded the case.(In the first 3 months, I used analogue type, 6 inch fluoroscope)
1 ml of contrast agent(Omnipaque 300, GE healthcare korea) per intervertebral foramen was injected and after that I classified the quality by 5 categories(Excellent, Good, Moderate, Poor) base on the fluoroscopic image.
## Loading required package: ggplot2
## Registered S3 methods overwritten by 'ggplot2':
## method from
## [.quosures rlang
## c.quosures rlang
## print.quosures rlang
##
## Attaching package: 'cowplot'
## The following object is masked from 'package:ggplot2':
##
## ggsave
Block Quality Classification
After contrast injection, Diluted 2% mepivacaine with NSS 20ml(0.1% mepivacaine) and hyaluronidase 1500IU was injected. I checked the clinical response on the next visit(after 3days ~ 7days), and classified the responses by 5 catergories(Excellent, Good, Moderate, No change, Aggravated).
I chose the linear-by-linear test for the association of block quality and clinical response. Statistical program was R.
| Variables | ||
|---|---|---|
| Sex | ||
| Male | 33 | |
| Female | 46 | |
| Age | ||
| Median | 59 | |
| Range | 37 - 81 | |
| Pain area | Frequency(%) |
|---|---|
| Upper arm | 28(35.4) |
| Lower arm | 23(29.1) |
| Neck | 21(26.6) |
| Shouder | 20(25.3) |
| Scapular | 19(24.1) |
| Hand | 15(19.0) |
| Inerscapular | 7(8.9) |
| Headache | 5(5.3) |
| Trapezius | 5(5.3) |
| Pain characteristics | Frequency(%) |
|---|---|
| Aching | 59(74.7) |
| Paraesthesia | 25(31.6) |
| Tightness | 9(11.4) |
| Tingling | 7(8.9) |
| Cramps | 3(3.8) |
| Decreased sense | 2(2.5) |
| Coldness | 1(1.3) |
| Pinrpicking | 1(1.3) |
| Weakness | 1(1.3) |
Total 79 patients were included in this study. Descriptive statistics expressed in graphical way. The p-value of asymptotic Linear-by-Linear association test was 0.0006, so there is a statitically significant association between block quality and clinical response.
Distribution of Sex and Age
Pain referring area and character
Balloon plotting for association
lbl_test(tab1)
##
## Asymptotic Linear-by-Linear Association Test
##
## data: Var2 (ordered) by
## Var1 (Poor < Moderate < Good < Excellent)
## Z = -3.41, p-value = 0.0006496
## alternative hypothesis: two.sided
It can be effective with T1/2 ro T2/3 transforaminal epidural approach for the cervical ventral epidural injection.