Identification of studies

The process of study inclusion is presented in Fig. 1. A total of 276 publications were identified through searching the databases, of which 186 remained after removal of duplicates and 176 were excluded by their titles and abstracts. The full-text of the remaining publications were assessed for eligibility. Two studies were excluded because they were conference papers or meeting proceedings [33, 34] and one was excluded because it was a case-series [10]. One publication was found when the references of relevant reviews and studies were assessed. Finally, eight studies were included in the meta-analysis.

Fig. 1

Preferred reporting items for systematic reviews and meta-analyses flow diagram

Overview of the included studies

General characteristics of the studies are summarized in Table 1. Dextrose prolotherapy (DPT) was compared with extracorporeal shockwave therapy (ESWT) in three studies [1, 4, 16]. Platelet-rich plasma (PRP) was the comparator in two studies [15, 16]. Only one study administered DPT along with exercise, in which the control group received saline injection plus exercise [13] and in another the comparator was plantar fascia and Achilles tendon exercise [14]. Dextrose prolotherapy was compared with corticosteroid injection in two studies [11, 16].

Table 1 Characteristics of the included studies

A total of 449 adult patients (mean age, 36.2–57.4 years) were evaluated in the included studies, with sample sizes ranging from 20 to 158 and plantar fasciitis symptom duration varying from eight weeks to 2.9 years. The concentration of the dextrose solution ranged from 1.5% [16] to 20% [4]. Dextrose was combined with anesthetics such as lidocaine and bupivacaine in all studies except for one [4]. The injections were performed under ultrasound guidance in all studies but two [12, 13]. The frequency of DPT ranged from one to three injections, 1–3 weeks apart. The needles used for injections were 22-, 25-, and 27-gauge and one study did not report needle specifications [16].

The shortest follow-up time was two weeks and the longest 36 months. Pain was evaluated in different studies using VAS, NRS, or the pain component of the FFI. Meanwhile, foot function was assessed using FFI [12,13,14,15], revised FFI (FFI-R) [16], AOFAS [13, 14], and FAAM [4, 11]. Plantar fascia thickness was also evaluated in three studies [4, 11, 12]. No adverse events or complications were reported with interventions in any of the studies.

Quality assessment

The results of the quality assessment are presented in Table 2. Of the eight studies, only one (1/8) had unclear risk of bias [13], while the rest (7/8) had high risk of bias based on Cochrane’s Collaboration tool [1, 4, 11, 12, 14,15,16]. All studies had low risk of reporting bias and random sequence generation [1, 4, 11,12,13,14,15,16]. The majority of studies had high risk of attrition bias [1, 4, 11, 12, 14, 15] and unclear allocation concealment [4, 11, 13, 14, 16].

Table 2 Risk of bias assessment by different items using the cochrane’s collaboration toolImmediate-term effects on pain

All the included trials reported the short-term effects of interventions on pain [1, 4, 11,12,13,14,15,16], while only six reported immediate-term [1, 4, 11, 12, 14, 16], and three long-term effects on pain [14,15,16]. Of the six studies reporting immediate-term effects on pain, Ersen et al. [14] and Mansiz-Kaplan et al. [12] showed significant immediate-term pain reduction with DPT compared to exercise and placebo, respectively. Overall, DPT was not superior to placebo/other non-surgical interventions for immediate-term pain reduction in plantar fasciitis (dppc2 = -0.46, 95% CI -1.37 to 0.45) (Fig. 2a).

Fig. 2

Forest plot of comparison: DPT vs placebo/other non-surgical interventions, outcome “pain”, a) immediate-term; b) short-term; and c) long-term

Short-term effects on pain

Pooled dppc2 showed significant large short-term effects of DPT compared to placebo/other non-surgical interventions on plantar fascia pain (dppc2 = -0.97, 95% CI -1.84 to -0.10) (Fig. 2b). Moreover, there was highly severe heterogeneity in short-term pain among the included studies (χ2 = 78.43, P = 0.00, I2 = 91.1%). Umay Atlas et al. showed that DPT was significantly more effective than placebo (saline) for pain reduction in the short-term [13]. Mansiz-Kaplan et al. also reported a significant short-term pain reduction with DPT compared to placebo [12]. Furthermore, Esrsen et al. illustrated similar results with DPT compared to exercise [14]. On the other hand, DPT was not superior to the control groups in this regard in other studies [1, 4, 11, 15, 16].

Subgroup analysis based on different control groups showed that DPT was only significantly superior to exercise and placebo for short-term pain reduction, while it was not better than PRP, corticosteroids, or ESWT in this respect (Fig. 3).

Fig. 3

Forest plot of comparison: DPT vs placebo/other non-surgical interventions by different control groups, outcome “short-term pain”

Long-term effects on pain

Of the three trials evaluating long-term effects on pain, only Ersen et al. reported significantly better long-term pain reduction with DPT than exercise. Also, the overall effect of DPT compared to the control groups was nonsignificant (dppc2 = 0.00, 95% CI -0.68 to 0.68) (Fig. 2c).

Immediate-term effects on foot function

Two trials used FAAM for the evaluation of foot function [4, 11], in which an increase in the total scores indicates improvement in foot function; therefore, they were not included in the meta-analysis for foot function outcome. Of the remaining six studies, immediate-term effects on foot function was reported in 4 [1, 12, 14, 16], short-term effects in all six, and long-term effects in three [14,15,16].

Of the four studies reporting immediate-term effects on foot function, Ersen et al. [14] and Mansiz-Kaplan et al. [12] showed significant immediate-term foot function improvement with DPT compared to exercise and placebo, respectively. Overall, DPT was not superior to placebo/other non-surgical interventions for immediate term foot function improvement in plantar fasciitis (dppc2 = -0.89, 95% CI -2.21 to 0.43) (Fig. 4a).

Fig. 4

Forest plot of comparison: DPT vs placebo/other non-surgical interventions, outcome “foot function”, a) immediate-term; b) short-term; and c) long-term

Short-term effects on foot function

Pooled dppc2 showed significant large short-term effects of DPT compared to placebo/other non-surgical interventions on foot function (dppc2 = -1.28, 95% CI -2.49 to -0.07) (Fig. 4b). Moreover, there was highly severe heterogeneity in short-term foot function among the included studies (χ2 = 47.50, P = 0.00, I2 = 93.2%). Umay Atlas et al. showed that DPT was significantly more effective than placebo (saline) for foot function improvement in the short-term [13]. Mansiz-Kaplan et al. also reported a significant short-term foot function improvement with DPT compared to placebo [12]. Furthermore, Esrsen et al. illustrated similar results with DPT compared to exercise [14]. On the other hand, DPT was not superior to the control groups in this respect in other studies [1, 15, 16].

Subgroup analysis based on different control groups showed that DPT was only significantly superior to exercise and placebo for short-term foot function improvement, while it was not better than PRP, corticosteroids, or ESWT in this regard (Fig. 5).

Fig. 5