Authors: Jose Alberto Pereira Pires, Ed Carlos Rey Moura, Caio Marcio Barros de Oliveira, Almir Vieira Dibai-Filho, Maria do Desterro Soares Brandão Nascimento, Plinio da Cunha Leal
Categories: 3800, chronic low back pain, conservative treatment, prolotherapy, Research Article, Clinical Trial/Experimental Study
Source: Medicine
Chronic low back pain (LBP) is defined as pain lasting longer than 3 months and is one of the conditions with the most significant social impact. Treatment is complex and includes proliferative agents used in prolotherapy. The mechanism is not known, but osmotic agents (hypertonic solutions of dextrose or glucose) cause cellular rupture and an inflammatory response that releases cytokines and growth factors that lead to scarring and reinnervation.
Patients with chronic LBP (>12 weeks) who were non-traumatic and unresponsive to at least 1 month of physical therapy were selected. All patients were followed up at a return visit at 1, 3, and 6 months for pain assessment using validated pain questionnaires and scales.
Nineteen patients were included in the conservative group and 19 in the glucose group, with the majority being women (57.9%). The patients were between 47 and 59 years of age (39.5%), mixed race (76.33%), married or in a committed relationship (73.7%), and had completed the study for 5.2 years. Overall mean body mass index was 27.3 ± 4.4 kg/m^2^ and was higher in the conservative group (28.0 ± 4.7 kg/m^2^). The groups showed differences in Visual Analog Scale scores, with median and amplitude values close to each other between the time points evaluated and increasing values in the glucose group, which had significantly higher values for this scale at the third evaluation (P = .031). When comparing the Rolland-Morris scale scores between the groups, there was a significant difference only in the 3-m assessment (P = .021). In the follow-up assessment, both groups showed significant improvement between T0 and the other assessment time points (P < .05) in all evaluations.
Both groups significantly improved on the evaluated scales during follow-up. Overall, no effects were attributable to the glucose components or the prolotherapy protocol.
Keywords: chronic low back pain, conservative treatment, prolotherapy
Chronic low back pain (LBP) is defined as pain lasting longer than 3 months and is one of the conditions with the most significant social impact. Approximately 80% of individuals experience LBP, and several risk factors, such as age, socioeconomic status, psychosocial factors, occupation, and obesity, promote the development of LBP.^[1,2]^ Furthermore, lower back pain ranges from 54% to 80%, resulting in an estimated socioeconomic impact of $50 billion annually in the United States.^[3]^
LBP is classified as acute, subacute, or chronic, depending on the duration of the episode. They lasted less than 6 weeks, 6 to 12 weeks, and >3 months, respectively. They are specific if the symptoms are caused by a pathophysiologic mechanism, such as a herniated disk involving the nerve root, inflammatory disease, infection, osteoporosis, rheumatoid arthritis, fracture, tumor, or nonspecific if the symptoms have no defined cause. The latter is the most common form, affecting 90% of people.^[4,5]^
Many instruments are used in research and clinical practice to assess pain, functionality, and quality of life, such as the Visual Analog Scale (VAS)^[6]^ and Roland Morris Disability Questionnaire,^[7]^ which assess the patient pain and functional capacity, respectively.
Treatment is complex and includes physical therapy, pharmacotherapy, interventional therapies, peripheral nerve injection, neuroablation or neurolysis,^[8]^ and proliferative agents used in prolotherapy. Back pain due to weakened ligaments can be alleviated by injecting proliferative solutions.^[9]^ Commonly used agents include irritants and chemotactic and osmotic agents.
Although the precise mechanism is not known, irritant agents (phenol, guaiac, and tannic acid) cause direct cellular damage, chemotactic agents (sodium morrhuate) cause an inflammatory response, and osmotic agents (hypertonic solutions of dextrose, glucose, glycerol, or zinc) cause cellular rupture and an inflammatory response that releases cytokines and growth factors that lead to scarring and reinnervation, resulting in tissue recovery.^[10]^
Hypertonic glucose 75% is considered a safe sclerosing agent.^[11,12]^ Complications include edema, short-term local urticaria (15 minutes), transient glycemic elevation in patients with insulin resistance, and short-term neurological alteration (2 hours) in patients with migraines and arrhythmias.^[11]^ Still, it has a high solute concentration, promotes the disintegration of the cells of the endothelial layer, and causes destruction and disintegration of the venous layer, which has a sclerosing effect on the walls of blood vessels.^[13]^ Because of its efficacy and safety, hypertonic glucose 75% is widely used as an organic agent. It rarely causes serious complications such as allergies, systemic reactions, and necrosis,^[13]^ and is not toxic.^[14]^
In the literature review by Dal Piva et al,^[15]^ most of the evaluated studies showed positive results regarding the effectiveness of prolotherapy. However, some studies showed short-term therapeutic action, while others demonstrated long-term effects.
Therefore, this study aimed to compare conservative treatment and prolotherapy with hypertonic glucose in patients with chronic LBP.
This randomized, blinded clinical trial was conducted between January 2020 and January 2022 in Pinheiro, Maranhão, Brazil, and performed with patients with lower back pain attended in the “Oficina da Coluna.” Data were collected by members of the Academic League of Neurology and Neurosurgery of the Baixada Maranhense, who were trained in the internal education of the league.
Patients aged >18 years with chronic LBP (>12 weeks) who were not traumatic and did not respond to at least 1 month of physical therapy were selected for the study. All patients with LBP with changes suggestive of facet degeneration on physical examination (pain on spinal extension and pain on palpation with facet joint topography) and tomography underwent facet block with 0.5 mL of 2% lidocaine. The same patients were reexamined after 2 hours, and those with more than 80% pain improvement were included in the cohort.
Individuals with root pain, signs of radiculopathy, or neurological deficits on physical examination during treatment were excluded from the study.
The study was conducted in accordance with the guidelines of the Declaration of Helsinki, approved by the Ethics Committee of the Federal College of Maranhão (CAAE:18311519.0.0000.5087), and registered in the Brazilian Clinical Trials Registry (REBEC) (RBR-543c9q9) under the universal test number (UTN) U1111-1252-5629. Informed consent was obtained from all participants.
Patients who met the inclusion criteria were randomly allocated into 2 groups using Randomizer® software on the day of the procedure. One group underwent conservative treatment, and the other underwent infiltration with a solution containing 1 mL/kg of hypertonic glucose 75%, containing Glucose Anidra (750 mg), water for injection (1 mL), Glucose Monohydrate (825 mg), into the facet joint for pain control, which was evaluated and followed up at the Outpatient Clinic (“Oficina da Coluna”) of the Macroregional Hospital, Pinheiro, MA.
The processes followed the CONSORT 2010 Statement.
The test and therapeutic blocks were performed at the Surgical Center of the Macroregional Hospital of Pinheiro, Pinheiro, MA, USA.
After randomization, patients in the conservative group underwent conservative treatment, which consisted of all clinical interventions (diet and lifestyle changes) that can be used to delay the worsening of the clinical manifestations. Patients in the hypertonic glucose group received 75% infiltration of hypertonic glucose (1 mL) into the facet joint and were informed of lifestyle modifications and encouraged to change their behavior.
All patients were initially assessed, received a pain diary, and were followed up at 1, 3, and 6 months for pain assessment using validated pain questionnaires and scales (VAS and Rolland Morris Disability Questionnaire). The lumbar facet joints were located between the L1-L2 and L5-S1 levels. The medial branches to be blocked were selected based on physical examination and imaging findings (computed tomography showing changes suggestive of facet degeneration).
A total of 42 patients with facet osteoarthritis detected by tomography of the lumbar spine were used for the sample size calculation. With a 95% confidence level and a 5% margin of error, the total sample size was 40 patients randomly divided into 2 groups (conservative × glucose 75%).
Data were analyzed using SPSS 21.0® (New York, EUA). Normality was tested using the Shapiro–Wilk test. The Mann–Whitney test was used to analyze numeric nonparametric variables, and the chi-square test was used to analyze categorical variables of the conservative and intervention groups. The Wilcoxon test was used to compare numeric variables during the follow-up (T0, 1, 3, and 6 months). Statistical significance was set at P < .05.
A total of 38 patients were 19 in the conservative group and 19 in the hypertonic glucose group, as shown in the CONSORT flowchart (Fig. 1).
Figure 1. Consort. São Luís, Maranhão, Brazil, 2022.
In the hypertonic glucose and conservative treatment groups, males (52.3%) and females (68.4%) predominated. In general, women predominated (57.9%), patients were between 47 and 59 years of age (39.5%), the mean age was 46.6 years, and the sample was predominantly multiracial (76.3%). Most patients were married or living in a committed relationship (73.7%) and had completed 5.2 years of study (92.1%). The mean overall body mass index was 27.3 ± 4.4 kg/m^2^ but was higher in the conservative group at 28.0 ± 4.7 kg/m^2^ (Table 1).
When comparing the glucose and conservative groups, differences were observed in the VAS, with the median and amplitude being similar between the time points assessed and increasing in the glucose group, which had significantly higher scores for this scale at the third assessment (3 months) (P = .031). In addition, when comparing the groups using the Rolland-Morris scale, a statistically significant difference was found only for the 3-m assessment (P = .021). At follow-up, both groups showed significant improvement between T0 and the other time points (P < .05) in all assessments (Table 2).
Thirty-eight patients were randomly divided into 2 groups (intervention with hypertonic glucose and conservative treatment) and followed up for 6 months, with a predominance of women between 47 and 59 years of age, with a mean age of 46.6 years, who were predominantly brown and married or in stable union and had 5.2 years of schooling. The mean body mass index was higher in the conservative group than in the control group.
The sociodemographic data from this study are consistent with those from studies that examined the influence of sex on LBP, such as that of DePalma et al,^[16]^ who showed that older women with obesity might have a higher prevalence of facet joint pain. A predominance of women was also observed in a retrospective study by Watson and Shay,^[17]^ who presented clinical outcomes of patients with chronic LBP, 64% of whom were women and had a mean age of 47.9 ± 12.2 years.
Female gender^[18]^ and advanced age^[18–20]^ are recognized as risk factors for facet arthrosis. Mariúba^[21]^ also found that 76.3% of the patients were female, and the mean age was 55.5 years. These data corroborate the results of this study.
In this study, both groups had high scores on the VAS, with the median within the classification range for severe pain, and on the Rolland-Morris Scale, with the median classified as indicative of physical disability.^[22]^
The VAS showed significantly higher scores in the glucose group at the third assessment (3 months) than in the glucose and conservative groups.
Studies^[23–25]^ using prolotherapy for chronic LBP had cohorts with similar baseline levels of pain and disability as those in our study. They were similar in using a proliferating agent, such as phenol, glycerol, glucose, or lidocaine, and had a follow-up duration of 6 months.
In a study by Yelland et al,^[9]^ significant reductions in pain intensity and disability scores from baseline were noted in all groups at 2.5 months through the end of the study, regardless of the intervention used. The data reported in this study also clarify the importance of conservative measures in maintaining the effectiveness of interventions in this population.
Shay^[17]^ study, conducted during a 1-year follow-up after using several prolotherapy agents, suggests that treatment may be effective for LBP, with a VAS improvement from 6.9 ± 2.1 before the intervention of 4.0 ± 2.9 at the follow-up. The contrasting data between this study and other studies may be related to the larger sample size and the data collection performed using convenience sampling.
When the groups were compared using the Rolland-Morris scale, a statistically significant difference was found in the 3-m assessment, with the glucose group having a more significant deviation and a higher pain level. Similar results were found in a study by Yelland et al,^[9]^ where no difference in the scale was found between the control and glucose groups.
At the follow-up assessment, both groups showed improvement between T0 and the other assessment time points, with no differences between the time points assessed. This was also observed in the study by Klein, Eek, and DeLong,^[24]^ in which the component groups subjected to the use of phenol/glycerol/glucose were tested against lidocaine-treated controls. Both groups showed significant improvements in the mean pain and disability scores at 6 months, but the differences were not significant.
Yelland et al^[9]^ reported that pain and somatic awareness improved in subjects who received 3 injections of hypertonic glucose or saline; however, the results did not differ significantly.
In contrast to these and our results, the study by Gül et al,^[25]^ conducted in patients with hip osteoarthritis, showed better results in the prolotherapy group (hypertonic glucose) than in patients subjected to an exercise-only protocol, which could lead to significant improvement in clinical outcomes and delayed surgery. There is evidence that prolotherapy may be more effective depending on the patient underlying disease. Studies have shown that prolotherapy stimulates collagen fiber production and strengthens the ligaments,^[26]^ which may be more effective in treating this type of disease.
In a review by Hauser et al,^[27]^ glucose prolotherapy was indicated for tendinopathies, knee and finger joint osteoarthritis, and spinal/pelvic pain due to ligament disorders, but it may not have the expected long-term effects.
The results of this study confirm the findings of the first clinical trial conducted on this topic,^[23]^ which concluded that the benefits of prolotherapy are temporary and generally not permanent unless the ligaments are strengthened. Strengthening the musculoskeletal system has been shown to lead to better long-term results because prolotherapy induces an inflammatory response that can lead to fibroblastic hyperplasia, and exercise stimulates the synthesis of the extracellular connective tissue matrix.^[28]^ This increases the strength of the ligamentous connection^[29]^ and induces the proliferation of fibroblasts to connect to the connective tissue, resulting in more effective clinical improvement.^[30]^ Therefore, prolotherapy alone cannot guarantee overall results unless accompanied by comprehensive multidisciplinary follow-up.
Another hypothesis proposed by Yelland et al,^[9]^ is that improvements in pain and disability markers may reflect only a reduction in the mean if the patients had severe pain at baseline. However, this response represents a natural return to pain levels, indicating the need for follow-up studies in patients with chronic pain.
Patients with LBP who remain symptomatic despite various therapeutic approaches are thought to have deficient ligament strength in the posterior elements of the sacroiliac joint, resulting in insufficient stability to allow effective muscle recruitment.^[31]^ This could have directly affected the results of this study, regardless of whether conservative treatment or prolotherapy was used.
In this study, facet blockade with 2% lidocaine (0.5 mL) was performed in all patients, effectively treating LBP from the beginning of follow-up. In the glucose group, prolotherapeutic behavior leading to reinnervation after the initial block may have resulted in recurrent pain, which was not observed in the conservative group.
According to previous studies,^[27,32]^ high glucose concentrations stimulate the growth and repair of normal cells and tissues and increase the production of growth factors and hormones.^[33]^ As with radiofrequency treatment, pain may return because of the destruction of the regenerating nerve fibers during denervation. Thus, reinnervation occurs, and new nociceptive inputs are formed in the zygoapophyseal joints.^[34]^ According to the prolotherapy assumption, this reinnervation process justifies the high VAS and Rolland-Morris scale scores in the glucose group in our study. However, these scores were higher and not significantly different from those in the conservative group.
Hebert et al^[35]^ stated that the recurrence of pain after neurolysis as an indication of joint reinnervation implies that there are still controversies regarding the procedure, such as weakness of the multifidus muscle (a stabilizer of the lumbar spine) due to its denervation, which could also explain the results of our study.
In their meta-analysis, Chung et al^[36]^ concluded that the evidence for the clinical benefit of hypertonic glucose prolotherapy in treating dense fibrous tissue injuries was insufficient. Imboden and Stone^[37]^ also concluded that there is no evidence that prolotherapy infiltrations are more effective than standard infiltrations already defined in the literature.
Imboden and Stone^[37]^ reported that lumbar spine exercises, aerobic exercises, reduction of excess weight, and patient education are effective in treating chronic LBP in most cases, which could explain the positive results at follow-up in the conservative group. However, no significant differences were observed in the glucose group.
Limitations include the lack of standardization of the most effective dose, and these values are not always available in other studies. In addition, the duration and frequency of follow-up may result in underreporting pain scores, and effective treatment may be complex without a multimodal and multidisciplinary approach.
This study concluded that experimental therapy is a safe and effective treatment for chronic LBP. However, prolotherapy is not more effective than conservative treatment. Overall, there were no effects on the glucose components or the prolotherapy protocol. Therefore, more studies on substances that promote prolotherapy are necessary since prolotherapeutics are more effective in the medium and long term, promoting fewer side effects, such as bone remodeling (osteoporosis), dyslipidemia, disorders related to glucose metabolism, and blood pressure changes, among others. Further clinical trials and meta-analyses should be conducted to verify which protocol is more effective in prolotherapy for LBP treatment of LBP.
Conceptualization: Jose Alberto Pereira Pires, Plinio da Cunha Leal.
**Data ** Jose Alberto Pereira Pires.
**Formal ** Jose Alberto Pereira Pires.
Investigation: Jose Alberto Pereira Pires, Plinio da Cunha Leal.
Methodology: Jose Alberto Pereira Pires, Plinio da Cunha Leal.
**Project ** Jose Alberto Pereira Pires.
Supervision: Ed Carlos Rey Moura, Caio Marcio Barros de Oliveira, Maria do Desterro Soares Brandão Nascimento, Plinio da Cunha Leal.
Validation: Ed Carlos Rey Moura, Caio Marcio Barros de Oliveira, Almir Vieira Dibai-Filho, Maria do Desterro Soares Brandão Nascimento, Plinio da Cunha Leal.
Visualization: Ed Carlos Rey Moura, Almir Vieira Dibai-Filho.
**Writing – original ** Jose Alberto Pereira Pires, Plinio da Cunha Leal.
**Writing – review & ** Ed Carlos Rey Moura, Caio Marcio Barros de Oliveira, Almir Vieira Dibai-Filho, Maria do Desterro Soares Brandão Nascimento, Plinio da Cunha Leal.
Ed Carlos Rey Moura, Email: edcrmoura@yahoo.com.br.
Caio Marcio Barros de Oliveira, Email: caiomboliveira@hotmail.com.
Almir Vieira Dibai-Filho, Email: almir.dibai@ufma.br.
Maria do Desterro Soares Brandão Nascimento, Email: m.desterro.soares@gmail.com.
Plinio da Cunha Leal, Email: pliniocunhaleal@hotmail.com.