Ultrasound Guidance in Paravertebral Injections of Oxygen-Ozone: Treatment of Low Back Pain

Low Back Pain (LBP) is a frequent and economically relevant disease with a cumulative life incidence of 70%. It is one of the most causative for long-lasting disability and the most common cause of invalidity in the population with an age >45 years [1-3]. Many studies demonstrated how patients affected by LBP have positive analgesic effects when treated by Oxygen/Ozone (O2/O3) infiltrations [4-6]. The administration of O2/O3 by paravertebral injections increases the threshold of pain through the nociceptive stimulation thanks to the engagement of the descending control of spinal nociception [7].


Introduction
Low Back Pain (LBP) is a frequent and economically relevant disease with a cumulative life incidence of 70%. It is one of the most causative for long-lasting disability and the most common cause of invalidity in the population with an age >45 years [1][2][3]. Many studies demonstrated how patients affected by LBP have positive analgesic effects when treated by Oxygen/Ozone (O 2 /O 3 ) infiltrations [4][5][6]. The administration of O 2 /O 3 by paravertebral injections increases the threshold of pain through the nociceptive stimulation thanks to the engagement of the descending control of spinal nociception [7].
The efficacy of ozone in the treatment of LBP is very well stated by Megalhães' analysis [4] but, as all mini invasive procedures, it may have side effects caused by the procedure itself as well as gas infiltration [7]. As a matter of fact, large volume of ozone in a single administration may have the following adverse or side effects [7][8][9]: acute muscular pain and deep visceral pain with high incidence of symptoms and exacerbation of the functio laesa, fatal air embolism due to erroneous injection of a big volume of gas in a blood vessel, burning and heaviness sensation, various complications caused by improper use of gas mixture (dosage and administration). The aim of our study was to evaluate the use of Ultrasound Guidance (USG) to practice O 2 /O 3 infiltrations in order to enhance better identification of the anatomical landmarks and therefore the need for a smaller volume of O 2 /O 3 .

Materials and Methods
The study was conducted at Pain Center of Policlinico Umberto I, University of Rome Sapienza, upon approval by the Institutional Ethical Committee: protocol number 718/12.
Sample size calculation was performed by calculator GraphPad InStat 3. We enrolled, from January 2014 to March 2015, 50 patients: 26 males and 24 females affected by LBP, not responsive to classical pharmacology, aged between 39 and 77 years, having a BMI range of 18.5-24.9.
Each patient was asked to sign an informed consent before starting the treatment. Patients were included in two groups: group U (n=25) and group AL (n=25). Randomization process was programmed by an online program (www.randomization.com) and statistical unit, in sealed envelope, was sent to the investigators for randomization purpose.
Exclusion criteria were: favism, pregnancy, severe cardiovascular diseases, hematological and respiratory failure, cancer and thyreopaties.
The treatment consisted in 10 administrations, once a week. Needle's caliber used for the treatment was 25 G * 90 mm in both groups.
In group AL, before the treatment, the mean of the VAS was 6.48, SD was 1.584, SEM was 0.3169, CI95%= [5.826-7.134]; the minimum was 4, the maximum was 10; normality test p value was 0.0032. After the treatment, the mean of the VAS was 3.04, SD was 2.508, SEM was 0.5016, CI95%=[2.005-4.075]; the minimum was 0, the maximum was 7; normality test p value was 0.0251 ( Figure 2).
The medians of VAS in group U and AL do not differ significantly. The two tailed p value is 0.1316 (Mann-Whitney Test).
The medians of discomfort rate of group U was 2.60 points less than group AL. Patients with unbearable discomfort is significantly different between the two groups: 0 in group U and 7 in group AL, tested by Fisher exact test (p<0.01) ( Figure 4).
The medians of discomfort in group U and AL differ significantly. The two tailed p value is 0.0008 (Mann-Whitney Test).

Discussion
Even if it's always more affirmed in many clinical practice with indisputable advantages in terms of efficacy and safety, international After located the trigger area by manual palpation, we proceeded to the injection of O2/O3 under USG at a distance of 2 cm from the spinal apophysis. The needle was inserted at 35-45° to the sagittal plane of the body. The correct position of the needle was confirmed by its identification near the articular facets thanks to the USG. After the injection, ozone should reveal as a hyperecogen area (Figure 1). The ozone administration was performed after negative aspiration test, in order to avoid blood vessels' involving.
Group AL underwent a single administration per side of 10 ml of O 2 /O 3 [O 3 20 mcg/ml]. The identification of the anatomical landmarks was only based on manual palpation. After located the trigger area, we proceeded to the injection of O 2 /O 3 at a distance of 2 cm from the spinal apophysis. The needle was inserted at 35-45° to the sagittal plane of the body. The ozone administration was performed after negative aspiration test, in order to avoid blood vessels' involving.
Pain assessment was performed as follows: -Pain intensity by VAS, before and after each treatment cycle (10 injections).
-Discomfort by a numeric scale from 0 to 10.
A binary variable was defined with discomfort. A discomfort ≥7 was classified as unbearable and a discomfort ≤ 6 as bearable. We defined as unbearable a discomfort which caused such a great burning and heaviness sensation that could threaten the continuity of the treatment and patient's confidence in the procedure.
A Medica-srl machine, model E100, was used to generate medical ozone (Ozonline International, Medica S.r.l., Via Sante Vincenzi, 48 -40138, Bologna, Italy) from oxygen and electricity. The machine converts medical oxygen into a mixture of O3 (0.05%) and O2 (99.95%) through an electrochemical process. It is equipped with a photometer, calibrated according to the classic iodometric titration of ozone, and a voltage system which regulates the concentration within a range from 5 to 100 µg/ml.

Results
In group U, before the treatment, the mean of the VAS was 6.44, SD was 1.294, SEM was 0.2587, CI95%=[5.906-6.974]; the minimum was 4, the maximum was 9; normality test p value was 0,0236. After the treatment, the mean of the VAS, was 2.22, SD was 1.958, SEM was 0.3916, CI95%=[1.392-3.008]; the minimum was 0, the maximum was 6 normality test p value was <0.0001 (Table 1).    That is why we decided to investigate if the use of USG in paravertebral injection of 5 ml O 2 /O 3 could help to reach the same results of 10 ml injections in terms of VAS decrease obtaining however the benefits of safety and low discomfort of 5 ml injection. For this purpose, this study allowed us to reconfirm the minimal volume, efficacious and with less discomfort, for USG injections.
The results showed that the discomfort caused by the procedure is significantly lower in group U than in group AL (p=0.0008) and the number of patients with unbearable discomfort is significantly lower in group U (p<0.01). A big volume injection is perceived as a heaviness and burning sensation by patients. Moreover, the reduced use of O 2 /O 3 volume increased the safety of this technique [7][8][9].
Both procedures were equal in term of VAS decrease. The medians of VAS in group U and AL do not differ significantly. The two tailed p value is 0.1316.
In order to benefit of the therapeutic effect of this gas mixture, O 2 /O 3 injection should be as closest as possible to the articular facet. But how to be certain of where we are injecting without the aid of USG?
In conclusion, we support the need for more investigations about USG paravertebral injections in the treatment of LBP to implement our preliminary results. In our opinion, the opportunity of a better precision and therefore, the use of a smaller volume of O 2 /O 3 makes this technique more accurate and safer.  Figure 3: Mean of the discomfort in group U and AL. The mean of discomfort was measured from 0 (no discomfort) to 10 (unbearable discomfort), the picture shows how the treatment was less painful in group U. The error bars show the SEM. In group AL, the discomfort rates have a double distribution: patients can be divided in those who did not show unbereable discomfort from the treatment (left side of the abscissa) and those who did (right side of the abscissa). Instead, discomfort rates of group U are more homogenous and all distributed on the left side of the abscissa: no one suffered of unbereable discomfort.
guidelines concerning paravertebral injections of O 2 /O 3 do not provide for USG. Our study demonstrates that also this kind of procedure takes advantages from the support of this device.
As showed by Megãlhaels' analysis, O 2 /O 3 therapy appears to yield positive results and low morbidity rates when applied percutaneously for the treatment of LBP [4].
Melchionda also reports positive effects of O 2 /O 3 therapy in terms of rapidity, simplicity and safety of the procedure in treating lumbar radiculopathies secondary to disc herniation [9].
Our clinical practice allowed us to understand that, when the anatomical landmarks were identified by palpation, patients underwent to 10 ml O 2 /O 3 [20 mcg] injections had better results in terms of VAS decrease than those underwent to 5 ml O 2 /O 3 [20 mcg] injections. Anyway they also had a greater discomfort caused by the big volume injection and some of them referred such a great burning and heaviness sensation that could have threaten the continuity of the treatment and patient's confidence in the procedure.