Mechanism/Animal Model Research

The in vitro mechanism of action of DPT has received little attention, and it is unclear how cellular changes associated with dextrose affect pain. Small concentrations of dextrose (15 mM, 0.27%) have been shown to alter the expression of various cytokines in multiple cell lines with less than 20 minutes of exposure, ^, while exposure to 0.6% dextrose in culture for 24 hours results in significant cellular death and dysfunction. ^, However, in vitro studies have not attempted to simulate the declining concentrations and limited exposure time seen in vivo.

Research using an in vivo rabbit model reported a proliferative effect of 10% dextrose in ligament tissue in the absence of an inflammatory response or cellular death. The same researchers confirmed organized connective tissue proliferation with 10% dextrose injections in three randomized controlled trials (RCTs) using the same rabbit model. All three studies demonstrated nearly a doubling of the ligament’s thickness after the dextrose injection, greater force was required to rupture the ligament, and histologic findings were normal. Studies have not compared the relative proliferative effect of differing dextrose concentrations in vivo. Although a proliferative effect of hypertonic dextrose injection has been documented, the therapeutic effects of hypertonic dextrose injection are likely multifactorial and a combination of proliferative and other biologic effects.

Clinical Research and Strength of Recommendations

Clinical research has been limited to relatively small studies. Despite modest sample sizes, numerous rigorous RCTs have reported a clinically meaningful effect of DPT in knee osteoarthritis (OA), temporomandibular dysfunction, rotator cuff tendinopathy, ^, lateral epicondylosis, ^, wrist pain, finger/thumb OA, ^, sacroiliac pain, hip OA due to hip dysplasia, Osgood-Schlatter disease, ^, Achilles tendinosis, and plantar fasciosis. ^,

Strength of recommendation (SOR) criteria are commonly used to summarize the cumulative status of evidence in a given area with level A or B SOR indicating likely benefit. DPT for knee OA has level A evidence. ^, In a clinical model of grade IV knee OA, exposure to intra-articular 12.5% dextrose injection was followed by the appearance of new, metabolically active, type I and II cartilage in post-treatment arthroscopies. While patients in this small case series improved clinically, the relationship between such improvement and apparent cartilage growth is not clear. Assessment of prolotherapy for several conditions meet B SOR criteria including temporomandibular dysfunction, rotator cuff tendinopathy, ^{, , ,} lateral epicondylosis, ^{, , , , ,} wrist pain, painful finger or thumb OA, ^{, , , ,} sacroiliac pain, Osgood-Schlatter disease, ^{, ,} and Achilles tendinopathy. ^{, ,} Treatment of plantar fasciosis has level B-C evidence. ^,

Training and Protocols

Over time, individual clinicians have refined the injection protocols and developed consensus-based guidelines. ^, Professional medical societies with a focus on prolotherapy, including the Hackett Hemwall Patterson Foundation (HHPF) and American Association of Orthopedic Medicine (AAOM), offer conferences and training worldwide. ( Table 9.1 ). HHPF has recently released “The Prolotherapy Procedural and Study Guides,” using an iterative consensus model.

Mechanism/Clinical Research

Neurogenic inflammation is characterized by an absence of leukocytes, and is due to peptidergic nerve fibers releasing pain-producing and degenerative neuropeptides. These neuropeptides, including calcitonin gene–related peptide (CGRP) and substance P, are primarily released by type C peptidergic nerve fibers. Clinical findings of allodynia and hyperalgesia are common due to alteration of C nerve fiber firing rates and thresholds. The rapid analgesic (not anesthetic) effect after subcutaneous 5% dextrose (D5W) injection over painful peripheral nerves suggests a neurogenic effect of dextrose.

In an RCT assessing caudal epidural injection of D5W versus normal saline in participants with back and buttock or leg pain, dextrose provided a significant analgesic effect within 15 minutes, lasting 48 hours or more, and this pattern was repeated after each of the 4 consecutive biweekly treatments. Proposed mechanisms for an apparent calming effect of dextrose in the presence of neurogenic pain include alteration of key cation channels or restoration of the normal cation-pump-maintained transmembrane potential by correction of relative neural hypoglycemia.

Multiple case series report pain reduction from subcutaneous dextrose injections over painful superficial cutaneous nerves in the knee, shoulder, elbow, Achilles tendon, and lumbar region, though assignment of causality is not possible in the absence of control. One randomized controlled trial has been reported using PIT in chronic Achilles tendinopathy, with injections about the Achilles tendon showing results comparable with eccentric lengthening exercises (ELE).

Training and Protocols

Neurogenic inflammation is common with dysfunction of superficial sensory or mixed motor and sensory nerves, ^, which can be located by palpation with training and injected with minimal risk using small-gauge needles. Training in PIT is largely outside of conventional medical education and is often taught without ultrasound guidance. Conference-based training was introduced by and remains available from content experts John Lyftogt and HHPF/IART. ^,

Mechanism/Clinical Research

A hypothesized mechanism of action for dextrose HD follows Bennett’s sciatic ligature model, where even mild constriction of a nerve caused neurogenic inflammation with sciatic nerve swelling, loss of axons distal to the ligatures, and neuroma formation at the level of the ligatures. A variety of injectates have been used, and saline, dextrose, corticosteroid, and platelet-rich plasma (PRP) HD injectates have been compared in the treatment of carpal tunnel syndrome (CTS). A favorable effect of HD using saline was reported in a study of CTS participants. Wu et al. reported that D5W outperformed saline and triamcinolone in treatment of CTS. PRP and D5W injection resulted in similar symptomatic benefit in a treatment-comparison trial in CTS.

Dextrose may offer advantages compared with other injectants. The risk of potential lidocaine toxicity is avoided with use of D5W or PRP without lidocaine. In addition, D5W HD without lidocaine does not result in nerve depolarization. In contrast, lidocaine depolarizes the nerve and renders distal fibers non-functional, limiting precise localization of the primary nociceptive source.

Training and Protocols

HD procedures will be discussed in this chapter and throughout this Atlas.

Head/Facial Region

Publications regarding the head/facial region have focused on DPT in or about the temporomandibular joint (TMJ). Three small RCTs with moderate to high bias compared the effect of dextrose to anesthetic injection in participants with painfully subluxing TMJs. ^{, ,} The protocols included injections of the TMJ capsule on multiple occasions in addition to an intra-articular injection. In all three studies, there was significant improvement in TMJ pain in both the dextrose and control injection groups, favoring dextrose upon meta-analysis. Despite a consistent reduction in frequency of perceived subluxation, a single-arm prospective study showed no change in measurable subluxation upon comparison of pre- and post-injection tomography of joint motion. The one complication observed in both the RCTs ^{, ,} and open-label studies was a significant reduction in mouth opening.

Louw et al., in a larger, low-bias, RCT, injected the TMJ intra-articular space and avoided capsular injection. They reported similar clinical benefit favoring dextrose over lidocaine injection in all types of TMJ dysfunction (myofascial, disc dysfunction, and arthritic), but the dextrose group showed improved mouth opening at the 3- and 12-month follow-up. The results argue for avoidance of capsular injection clinically and in future study protocols. ^,

Cervical Spine/Neck

DPT was studied in a small group of participants ( n = 6) with cervical laxity with greater than 2.7 mm of cervical translation on flexion-extension x-rays and referred shoulder pain. Using fluoroscopic guidance, subjects underwent injections with 12.5% dextrose into the involved posterior elements, lamina, and spinous processes. Reduced cervical translation in flexion-extension x-rays, along with a significant reduction in pain, was observed. For facet-mediated pain, Hooper et al., in a small case series ( n = 18) involving 20% dextrose injection into the zygapophysial joint at 2 to 4 levels using fluoroscopic guidance in post whiplash patients, found DPT improved pain and function.

Thoracic Spine/Upper to Mid Back

In a study of 70 subjects with chronic axial spine pain treated with 20% DPT targeting the ligaments and facet joint capsules at symptomatic levels, Hooper et al. reported clinically significant improvement as measured by the patient-specific functional scale (PSFS). Of interest is that 50 of the 70 participants were litigants, with outcomes comparable to non-litigants.

Shoulder Girdle

DPT for rotator cuff tendinopathy has been the object of considerable research. Bertrand et al. conducted a three-arm blinded RCT ( n = 73) comparing landmark-guided injection with dextrose into the painful enthesis to lidocaine at the enthesis and superficially. Seven et al., in an open label RCT ( n = 101), compared ultrasound-guided dextrose injections into the subacromial bursa and tender entheses to extensive physiotherapy. While both groups had significant improvement from baseline measures, the dextrose group significantly outperformed the control at 6 and 12 weeks in pain score reductions, Western Ontario Rotator Cuff (WORC) index, and Shoulder Pain and Disability Index (SPADI).

In a pairwise and network meta-analysis by Lin et al. comparing all injection therapies in the treatment of rotator cuff tendinopathy, platelet-rich plasma and prolotherapy injections yielded better outcomes than control injections in the long term (>24 weeks). Study heterogenicity limited conclusions to level B confidence. A systematic review by Catapano et al. of DPT for symptomatic rotator cuff tendinopathy concluded that dextrose demonstrated at least short-term improvements in pain and function compared to physical therapy alone. There was a high risk of bias and variable efficacy within the studies, but repeated multisite injections showed more consistent improvements.

Elbow

Rabago et al., in a three-arm RCT, compared prolotherapy with dextrose and dextrose-morrhuate injections to watchful waiting for lateral epicondylosis. Patient-Rated Tennis Elbow Evaluation scores at 16 weeks in both prolotherapy injection groups were significantly better than in the watchful waiting group, and there was no difference between the dextrose and dextrose-morrhuate injection groups. Bayat et al. compared DPT to methylprednisolone. While both injections resulted in significant clinical improvements at 3-month follow-up, dextrose was superior in Quick DASH (Disabilities of Arm, Shoulder and Hand) and pain improvements. In the most rigorous study of dextrose injection therapy for lateral epicondylosis, Yelland et al. conducted a single-blind RCT ( n = 120) comparing 20% DPT, physiotherapy, and combined treatment with prolotherapy and physiotherapy. At 1 year, all groups showed significant improvement with no significant differences between groups. Dong et al., in a systematic review and Bayesian network meta-analysis of all injection therapies for lateral epicondylalgia, concluded that botulinum toxin, platelet-rich plasma, autologous blood injection, hyaluronate injection, and prolotherapy can be considered for lateral epicondylitis, but cortisone was not recommended. Hyaluronic acid injection and prolotherapy might be superior, but Dong and colleagues suggested additional research is needed.

Wrist/Hand

Reeves et al. compared hypertonic dextrose (10%) solution to lidocaine alone injection for proximal interphalangeal (PIP), distal interphalangeal (DIP), and trapeziometacarpal (TMC) OA at 0, 2, and 4 months. There was no significant difference at rest, but a significant improvement in pain with finger function and finger flexion range of motion in the dextrose group was reported at 6-month follow-up. Jahangiri et al. in an RCT ( n = 60) compared 3 monthly dextrose injections for TMC OA to a group with saline (at 0 and 1 months) and methylprednisolone (at 2 months). At 6-month follow-up, pain and pinch improvements were significantly more in the dextrose group. The above studies were included in two systematic analysis of RCTs for the treatment of knee and hand OA. ^, Krsticevic et al. in a narrative review concluded that there was limited evidence indicating a beneficial effect of prolotherapy for finger/thumb OA management, and Hung et al. in a systematic review and meta-analysis concluded that dextrose injection reduced pain in patients with hand OA.

At the wrist, Hooper et al. in an RCT ( n = 39) compared periscaphoid and perilunate injections with hypertonic dextrose to lidocaine in participants with dorsal wrist pain and normal x-rays. At 3 months there was no difference in outcomes, but at 12 months the Patient Rated Wrist Evaluation (PRWE) scores were significantly improved in the DPT group compared to the lidocaine group. Both groups improved more than the minimal clinically important difference (MCII) for PRWE, indicating a clinical benefit from lidocaine needling as well.

Wu et al. have reported three RCTs on HD for CTS. One trial ( n = 34) compared saline HD (5 mL) into the intracarpal region to superficial injection of saline outside the carpal tunnel, and found a therapeutic benefit to HD in mild-to-moderate CTS. A second study ( n = 49) compared a single HD with 5 mL of D5W to HD with saline around the median nerve, and showed the D5W group had a significant reduction in pain and disability, improvement on electrophysiologic response measures, and decreased cross-sectional area of the median nerve compared to the saline HD control group. A third study ( n = 54) compared a single HD with 5 mL of D5W to triamcinolone injection 3 mL, and showed a significant reduction in pain and disability compared to triamcinolone. Shen et al., in an RCT ( n = 52) comparing HD with dextrose to HD with PRP for the treatment of mild to moderate CTS, reported significant and comparable improvements in pain and functional status in both groups.

Studies by Wu et al. are among the most rigorous RCTs for any dextrose-related therapy and strongly support an independent biologic action for dextrose D5W HD compared to saline or corticosteroid injection.

Lumbosacral Area

Yelland et al. assessed the efficacy of needling the entheses of symptomatic lumbo-pelvic ligaments and sacroiliac joints with dextrose versus lidocaine ( n = 110). Both groups reported a clinically significant improvement in pain and disability at the 2-year follow-up irrespective of the solution injected, suggesting that needling of entheses has a therapeutic effect. These findings are consistent with other DPT studies with a needling control, ^, and imply that needling controls should not be considered as a control in future research design. Instead, inclusion of a non-injection control should be considered.

Maniquis-Smigel et al. in a prospective uncontrolled cohort study ( n = 32) evaluated the effect of caudal epidural injection of D5W (without an anesthetic or corticosteroid component) in participants with chronic low back pain radiating to the gluteal area or lower extremities. Participants received 10 mL D5W caudally biweekly (x4) and then as needed. At the 1-year follow-up, 66% of participants reported a 50% or more reduction in chronic pain, and improved disability ratings on Oswestry Disability Index.

Kim et al. in a single-blind RCT ( n = 48) compared 25% dextrose to triamcinolone injections into the sacroiliac joint using fluoroscopic guidance with up to three biweekly injections. At 15-month follow-up, dextrose injection showed a significant improvement in pain compared to cortisone, with more subjects reporting a ≥50% reduction in back pain (58.7% vs. 10.2%). For coccygodynia, Khan et al. published the largest case series with 37 consecutive patients with > 6 months of pain treated with 25% DPT at the area of maximum pain. There was a reduction in VAS pain score in excess of 70% after 2 injections, but only 2-month follow-up was reported. Thirty patients had good relief and seven had minimal or no relief.

Hip/Pelvic Area

There are no randomized trials of DPT for primary hip OA. Gul et al. in an open-label RCT treated 41 patients (46 hips) with hip OA secondary to developmental dysplasia with hypertonic dextrose injections to tender periarticular tendons at 21-day intervals for a maximum of 6 injections versus 30 sessions of supervised progressive resistance training. DPT outperformed the exercise control for improvements in Harris Hip Score (HHS) and pain scores at the 6-month and 1-year follow-ups. These improvements were clinically significant, exceeding twice the minimal clinically important difference (MCID) for VAS pain improvement and above the MCID for HHS.

DPT has been reported as a treatment for athletic pubalgia. Topol et al. reported on a case series of 72 consecutive elite soccer or rugby athletes with chronic adductor origin or pubic symphysis pain. Subjects received 3 monthly treatments of palpation-guided 12.5% DPT, and 66 of 72 (92%) returned to full sport within 3 months.