Patients with Ehlers-Danlos Syndrome Experience Reduced Effectiveness of Lidocaine Local Anesthetic: A Randomized Cross-Over Clinical Trial

Abstract

Background:

The Ehlers-Danlos syndromes (EDS) are a collection of heritable connective tissue disorders. Local anesthetic resistance is commonly reported by patients with EDS, but there is little clinical evidence to support this finding. We hypothesized that patients with EDS would have increased anesthetic resistance using lidocaine compared to healthy controls.

Methods:

An interventional study design was used. Participants (total n=135: EDS n=91, healthy n=44) completed a series of sensation tests using a standardized 10g monofilament following subcutaneous injection with saline or lidocaine at two separate sites on the right forearm in a randomized order. Participants rated their sensation at each site relative to non-anesthetized skin. Sensation was used as a surrogate measure of anesthetic effect. Participant ratings were converted into two measures - a dichotomous variable to represent adequate anesthesia, and a delta sensation score. This study was registered on clinicaltrials.gov (NCT05603741).

Results:

Delta sensation scores were not different at five minutes post-injection between EDS and healthy participants (p=0.2), but were significantly different at 15 minutes (1 [0, 2] a.u. vs. 2 [1, 2] a.u; p=0.002) and 30 minutes (1 [0, 2] a.u. vs. 2 [1, 2] a.u.; p=0.003) post-injection. There was no difference in categorical “anesthesia” at five minutes between EDS and controls (p=0.2). A smaller proportion of EDS experienced “anesthesia” at 15 (60% vs. 84%, p=0.006) and 30 (44% vs 91%, p=0.003) minutes post-injection compared to healthy controls.

Conclusion:

Patients with EDS experienced increased sensation with lidocaine, suggesting shorter duration of effect of lidocaine local anesthetic. This evidence validates previous patient reports and small case studies suggesting this phenomenon. Healthcare providers should be aware of this reduced effectiveness experienced by EDS patients, to adjust care accordingly.

Introduction

The Ehlers Danlos Syndromes (EDS) are a heterogeneous group of heritable connective tissue disorders.¹ There are currently 13 classified subtypes, with hypermobile EDS (hEDS) being the most common type.¹ EDS is characterized by joint hypermobility, skin hyperextensibility, and skin fragility.¹ Ongoing concerns regarding the clinical recognition of EDS and patient care can significantly affect patient quality-of-life.² Local anesthesia, which is a mainstay of many dental and minor surgical procedures to temporarily numb pain, has been reported to be inadequate for patients with EDS.³

There have been consistent reports of inadequate analgesia experienced by patients with EDS. In 2003, Hakim and Grahame found that 58% of joint hypermobility patients (n=172) indicated they felt a local anesthetic injection was not as effective as it should have been, compared to only 21% of healthy participants.⁴ In 2019, Schubert et al. found that 88% of EDS patients (n=988) reported they had issues with local anesthetics working improperly, compared to only 33% of healthy participants.⁵ Multiple case reports also describe EDS patients experiencing resistance to local anesthetics.^6,7 Despite these reports, the current objective evidence of local anesthetic resistance in EDS is of low quality. Only one experimental study has previously evaluated the effectiveness of local anesthetic in EDS (n=8) and healthy controls (n=8). This small study used intradermal lidocaine to evaluate anesthetic resistance.⁸ Although anesthesia was effective at 5 minutes post-administration in both groups, no EDS patients experienced adequate anesthesia at 1 hour post-administration, compared to 75% of controls.⁸

This study investigated the effectiveness of lidocaine, a commonly used local anesthetic, in EDS patients compared to a healthy population to provide more thorough empirical data regarding the rate of local anesthetic effectiveness with lidocaine in EDS patients. We hypothesized that EDS patients would experience lower lidocaine anesthetic efficacy compared to healthy participants.

Methods

Study Participants:

Patients with EDS and healthy controls (HC) were recruited for this study at the 2019 Ehlers-Danlos Society Global Learning Conference in Nashville, Tennessee (n=88 EDS, n=20 HC) and at the University of Calgary in Calgary, Alberta (n=3 EDS, n=24 HC). Patients reported a physician diagnosis of EDS based on the 2017 criteria.¹ For patients with hEDS, this was the new 2017 hEDS criteria, and for patients with other types of EDS, this was genetic testing. If they suspected that they had EDS but had not previously been formally diagnosed, then they underwent a focused history and physical examination to determine diagnosis of EDS using the 2017 diagnostic criteria and goniometer measurements. This applied to n=1 participant from the Calgary site and no participants from the Nashville site. Healthy participants had no diagnosis of EDS. All participants were 18 years of age and older. Individuals with a known allergy to lidocaine were excluded from the study.

Overall Study Design:

This study received ethical approval from the Conjoint Health Research Ethics Board at the University of Calgary (REB22-1157) and the Institutional Review Board at Vanderbilt University Medical Center (IRB#190957). This study was registered on www.ClinicalTrials.gov on July 26, 2019 at Vanderbilt University Medical Center (ID: NCT04036305) and at the University of Calgary ((ID: NCT05603741) on November 10, 2022. Study enrollment through Vanderbilt University Medical Center began on July 30, 2019 and the University of Calgary on November 10, 2022. Participants from Vanderbilt University Medical Center were recruited by self referral at the Ehlers Danlos Society 2019 patient conference. At the University of Calgary we identified 72 potential participants for participation in the study. All participants provided written informed consent to participate in the study. An interventional study design was used, where both study groups underwent the same study intervention and were compared. Unless a prior diagnosis was already established, participants underwent a focused history and physical examination to determine diagnosis of EDS using the 2017 EDS diagnostic criteria. Studies were conducted with two members of the research team, one who performed the injection and another blinded assessor who performed the sensation testing. There were two standardized injection sites on the forearm (in horizontal arrangement across the arm) and the agent injected in each site (lidocaine or saline) was randomized upon enrollment.

Study Day:

On the study day, participants were asked a series of questions to determine diagnosis of EDS. Then the right anterior forearm was prepared for injection, labelling sites with a surgical marker (Figure 2). A control site, where no injection would occur, was labelled at least 1.5 inches distally from the antecubital fossa. The two testing sites were labelled at least 2 inches below the control site, ensuring the test sites were 2 inches apart on the same horizontal axis. This was done to avoid the risk of a proximal injection affecting a distal assessment site. Slight adjustments to the placement of the testing sites were occasionally made to avoid injecting into a vein. Prior to injection, the study team performed sensation testing to provide a baseline for participants.

Figure 2.

Diagram of injection sites on the participant’s right arm. A control site (red) was labelled at least 1.5 inches distally from the elbow joint. The two testing sites (blue) were labelled at least 2 inches below the control site, ensuring the test sites were 2 inches apart on the same horizontal axis. Slight adjustments to the placement of the testing sites were occasionally made to avoid injecting into a vein. This figure was created with BioRender.com.

Participants were injected subcutaneously in their right forearm with 0.5mL of 2% lidocaine without epinephrine (Teligent Canada, Mississauga, ON) at one site, and 0.5mL of 0.9% saline (Pfizer Canada Inc, Kirkland, QC) at the other site, in a randomly assigned order. Both the participant and study personnel administering the sensation testing were blinded to the drug-location assignment. Sensation testing occurred at five (5M), fifteen (15M), and thirty (30M) post-subcutaneous injection. A 10-gram monofilament (Baseline Evaluation Instruments, White Plains, NY) was used for each testing site. This monofilament is standardized to bend at 10g of pressure, providing a consistent amount of force across multiple trials and operators. The participant was asked to look away from their forearm during sensation testing to minimize confounding factors. The 10g monofilament was first applied to the control site and then to one injection site. The monofilament was pressed three times at each site to the point where the monofilament bent slightly to provide the same pressure at each site. The participant was asked to rate their sensation at the test site compared to the control site on a scale of zero to three; zero meaning the sensation was absent, one meaning the sensation was dull but not sharp, two meaning the sensation was sharp but less than the control site, three meaning the sensation was the same as, or worse than, the control site. This was repeated for the second injection site. This sensation score was used as a surrogate measure for anesthetic effect.

After the study, participants were emailed a REDCap survey^9,10 with questions regarding demographics, history of local anesthetic resistance, and EDS comorbidities (if applicable).^9,10 Some participants did not complete the survey, leading to missing demographic variables in Table 1. The entire study duration was approximately one hour.

Table 1:

Participant demographics categorized by patient diagnosis.

	EDS (n=91) n (%)	HC (n=44) n (%)	P - Valuea
Sex *			0.01
Male	5 (5)	9 (20)
Female	85 (93)	35 (80)
Race †			0.004
White	76 (94)	32 (74)
Other	5 (6)	11 (26)
Age (years)			0.45
Mean (SD)	37 (14)	41 (17)
Min-Max	18-69	18-71
Median (25th, 75th percentile)	34 (25-47)	37 (24-57)
Weight (kg)			0.45
Mean (SD)	73.6 (18.3)	75.3 (16.0)
Min-Max	43-123	48-115
Median (25th, 75th percentile)	71.8 (60-82)	71.4 (63-82)
Height (cm) †			0.07
Mean (SD)	162 (18)	165 (18)
Min-Max	67-183	62-188
Median (25th, 75th percentile)	163 (160-170)	167 (163-171)
Body Mass Index (BMI) †			0.89
Mean (SD)	31 (21)	30 (22)
Min-Max	17-177	18-167
Median (25th, 75th percentile)	26 (22-30)	26.6 (23-30)
Body Surface Area (BSA) †			0.45
Mean (SD)	1.8 (0.3)	1.8 (0.2)
Min-Max	1.0-2.4	1.1-2.4
Median (25th, 75th percentile)	1.8 (1.7-1.9)	1.8 (1.7-2.0)
Have had local anesthesia †			0.049
Yes	80 (99)	39 (91)
No	1 (1)	4 (9)
Had a problem with local anesthetic working properly †			<0.001
Yes	71 (89)	4 (10)
No	9 (11)	35 (90)

^{*/ †}These variables were participant self-reported or calculated using self-reported values.

^*There is 1 missing value.

^†There are 11 missing values.

^a:Fischer’s Exact Tests used to compare categorical variables and Mann-Whitney U Tests used to compare continuous variables.

Statistical Analysis:

Data were exported from REDCap^9,10 and imported into IBM SPSS Statistics 29 (IBM, Armonk, NY) for analyses. Participants’ sensation scores were converted into two variables for statistical analysis. First, scores were converted into a “delta sensation score” (sensation score at saline location [compared to control location] - sensation score at lidocaine location [compared to control location]). The delta sensation score assesses incrementally increased effectiveness of lidocaine compared to saline. Delta sensation scores equal to zero indicated no anesthetic effect, delta sensation scores equal to three indicated the patient experienced full anesthetic effect, and negative delta sensation scores indicated that saline had more of an anesthetic effect than lidocaine. Second, scores were converted in a dichotomous fashion to “anesthesia” (if sensation score with lidocaine was less compared to sensation score with normal saline) or “no anesthesia” (if sensation score with lidocaine was greater than or equal to sensation score with normal saline). The distribution of each variable was tested using Shapiro-Wilk normality tests. The primary analysis was delta sensation score at 5M post-injection in EDS compared to HC. Secondary analyses included comparisons at 15M and 30M post-injection, comparisons of “anesthesia” at each time point between EDS and HC, comparisons excluding participants who reported anesthetic effects from saline, and comparisons by biological sex. The analyses of the categorical outcome of “anesthesia” were tested using a Fisher’s Exact test. The analyses of the continuous delta sensation scores were tested using Mann-Whitney U tests. Demographic variables were compared between EDS and HC using Fisher’s Exact Tests and Mann-Whitney U tests for categorical and continuous data, respectively. Analysis of covariance (ANCOVA) was performed to adjust to weight differences in healthy controls. Data are presented as median (25^th, 75^th percentile) for continuous variables and number and/or percentage of participants for categorical variables.

Sample Size Calculation:

Based on a previous patient survey⁵, 88% of EDS patients and 33% of non-EDS patients reported prior local anesthetic resistance at some point in their lives. We expected that both proportions would be lower during our one-time testing. We estimated that the rate of local anesthetic resistance is 65% in the EDS patients and 40% in the healthy participants. Thus, a sample size of 87 EDS patients and 44 HC, would give us 80% power with an alpha error rate of 0.05 to show this difference¹¹.

Results

Demographics:

A total of 135 participants completed this study. Of these, 91 were participants with EDS (93% female; age: 34 [25, 47] years), and 44 were HC (80% female; age: 37 [24, 57] years, Figure 1). Within the subset of participants with EDS, 96.7% (n=88) had the hypermobile type (hEDS), 1.1% (n=1) had the classical type, 1.1% (n=1) had the vascular type, and 1.1% (n=1) had a non-classified type. Participant demographics are shown in Table 1.

Figure 1.

CONSORT flow diagram. A total of 135 participants were enrolled in the study.

Delta Sensation Scores:

At 5M post-injection, the median delta sensation scores of EDS patients (0 [0, 1] arbitrary units [a.u.]) and HC (1 [0, 2] a.u.) were not significantly different (p=0.15; Figure 3a). The median delta sensation scores, indicating amount of anesthesia, were lower for patients with EDS than HC at 15M, (1 [0, 2] a.u. vs. 2 [1, 2] a.u; p=0.002) and 30M (1 [0, 2] a.u. vs. 2 [1, 2] a.u.; p=0.003) minutes post-injection. Raw sensation scores are shown in Table S1.

Figure 3.

A. Boxplots of median delta sensation scores for EDS and HC at 5 minutes, 15 minutes and 30 minutes post-injection. The box spans from the 25^th -75^th percentile. The dots represent individual participant outliers. A higher delta sensation score reflects a higher level of anesthesia. B. Percentage of participants with Ehlers-Danlos syndrome (EDS) and healthy controls (HC) who experienced “anesthesia” at 5 minutes, 15 minutes and 30 minutes post-injection.

Lidocaine Effectiveness:

At 5M post injection, the proportion of EDS compared to HC that experienced “anesthesia” was not significantly different (47% vs 59%; p=0.21; Figure 3b). Fewer EDS than HC experienced “anesthesia” at 15M (60% vs 84%; p=0.006) and 30M (53% vs 80%; p=0.003) post-injection.

“Over-responders” who experienced anesthesia from saline:

When excluding participants who experienced some anesthesia with saline (n=44), there was no statistically significant difference between the proportion of EDS patients and HC who experienced “anesthesia” at 5M (51% vs 68%; p=0.17) and 30M (56% vs 79%; p=0.059), but there was a statistically significant difference at 15M (60% vs 86%; p=0.027). The median delta sensation scores of EDS patients were lower than HC at all time points (5M: 1 [0,1] a.u. vs 1 [0, 2] a.u., p=0.033; 15M: 1 [0, 2] a.u. vs. 2 [1,3] a.u., p <0.001; 30M: 1 [0, 2] a.u. vs 2 [1, 2] a.u., p=0.025).

Role of biological sex:

When looking at female participants (n=120), there was no significant difference between the proportion of EDS patients and HC who experienced “anesthesia” at 5M (48% vs 66%; p=0.11), but there was a significant difference at 15M (61% vs 94%; p<0.001) and 30M (52% vs 89%; p<0.001; Figure 4a). There was no significant difference between median delta sensation scores of EDS patients and HC at 5M (0 [0, 1] a.u. vs. 1 [0, 2] a.u.; p=0.11), but a significant difference at 15M (1 [0, 2] a.u. vs 2 [1, 2] a.u.; p=0.001) and 30M (1 [0, 2] a.u. vs 2 [1,2] a.u.; p<0.001; Figure 4b). When looking at male participants (n=15) there was no significant difference between the EDS population and HC at any time points, with both analyses (Figure 4c-d). Male healthy controls weighed more and had a higher BMI than the female controls (Table S2). Male healthy controls had a lower percentage of adequate anesthesia compared to female healthy controls at 15M (p<0.001) and 30M (p=0.01), but not 5M (p=0.1). After adjusting for body weight, there were no differences in delta sensation scores between male and female healthy controls (Table S3).

Figure 4.

A. Percentage of female participants with Ehlers-Danlos syndrome (EDS) and healthy controls (HC) who experienced “anesthesia”. B. Median delta sensation scores for female EDS and HC. A higher delta sensation score reflects a higher level of anesthesia. C. Percentage of male participants with Ehlers-Danlos syndrome (EDS) and healthy controls (HC) who experienced “anesthesia”. D. Box plots of median delta sensation scores for male EDS and HC. The box spans from the 25^th -7^th percentile. The dots represent individual participant outliers. A higher delta sensation score reflects a higher level of anesthesia.

Participant Survey:

Overall, 89% of EDS (n=81) and 98% of HC (n=43) completed the participant survey. Most EDS (99%, n=80) and HC (91%, n=39) had received local anesthetic for a minor surgery or dental procedure at least once prior to the survey (compared to HC, p=0.049; Table 1). The overwhelming majority of EDS reported that they previously had a problem with local anesthetic working properly (89%, n=71) compared to only 10% (n=4) of HC (p<0.001).

Discussion:

We have demonstrated that individuals with EDS experience decreased anesthesia in response to local anesthetic, compared to HC, using subcutaneously injected lidocaine. A larger proportion of participants with EDS experienced inadequate analgesia compared to the HC at 15M and 30M post-injection, with increased sensation scores post-local-anesthesia. The relative efficacy of local anesthetic in EDS patients was lower than in HC, demonstrating that EDS patients were experiencing decreased local anesthetic effectiveness with lidocaine. These results are consistent with prior patient reports suggesting decreased effectiveness exists in EDS.^2-8,12 Importantly, most participants with EDS in this study had hypermobile EDS and, therefore, the study results may not be fully reflective of anesthetic effects in other types of EDS.

Potential mechanisms of decreased lidocaine local anesthetic effectiveness:

Of note was the lack of a significant difference between EDS and HC at 5M post-injection and the initial effectiveness was similar between groups. After 5 minutes, the effectiveness of the lidocaine local anesthetic increased in HC, while it remained ineffective in EDS patients throughout the 30 minutes of testing. While this might suggest reduced effectiveness due to an inherent difficulty with how the anesthetic is acting in patients with EDS, it is more likely to represent a faster clearance of lidocaine in patients with EDS compared to healthy controls. One prior study that suggested a more rapid dispersal of local anesthetic in EDS evaluated patients over one hour.⁸ Conversely, our results could be explained by a delayed onset of local anesthetic effectiveness, rather than a lack of local anesthetic effectiveness. Other theories regarding the mechanism of local anesthetic resistance in EDS patients, including a potential molecular mechanism such as a genetic mutation that impacts responsiveness or pain sensitization.^5,13 In the present study, however, genetic testing was not conducted and we were not able to test these theories. It is also important to note that the genetic etiology of hypermobile EDS is not currently understood, and the specific genetic mutation responsible is not known. Future studies could evaluate LA response over longer time durations. Further, only one local anesthetic dose was tested in this study. Some have speculated that a molecular mechanism may drive anesthetic resistance in EDS, in which case the phenomenon could not be overcome by increasing the local anesthetic dose.¹³ The effectiveness of different anesthetic doses at overcoming resistance could be investigated in future research.

Sex Differences:

There is minimal evidence detailing sex differences in local anesthetic effectiveness.¹⁴ Small studies have demonstrated no sex difference in local anesthetic effectiveness between male and female patients.^14,15 The majority of patients with EDS are female¹⁶, and this is reflected in the primarily female demographic of this study. The trends for female patients match the overall trends of the study. There were no significant differences detected among male participants, but this is likely complicated by the smaller sample size of male patients and consequently underpowered analyses. Additionally, the male healthy controls were taller and weighed more than the female healthy controls. This could also contribute to the reduced percentage of anesthetic effectiveness in male healthy controls compared to female healthy controls, as once correcting for body weight, there were no significant differences in delta sensation scores amongst the healthy controls. There were no significant differences between male and female EDS patients.

Excluding saline-anesthetic individuals:

Some participants reported experiencing an analgesic effect with saline; this was likely due to confounding variables such as the lidocaine spreading near the saline testing site, or issues with patient perception. It could, however, be due to true anesthetic effect from saline.¹⁷ When omitting these participants, the trends remain similar. A larger proportion of EDS patients experienced decreased lidocaine effectiveness at 15M, but there was no significant difference at 30 minutes, which could be because the study was not powered for this subgroup analysis.

Limitations:

Patient-completed questionnaires were used to collect demographics including height, biological sex, gender and race. Incomplete questionnaires resulted in data missing for gender (n=107). Due to the missing gender demographics, the study was unable to comment on gender-based differences. In this study most participants had hEDS, the most common type of EDS, limiting our ability to comment on other EDS subtypes. Future research should investigate the effects of local anesthetic with different subtypes. Since EDS primarily affects females¹⁶, it was difficult to recruit enough male participants to make meaningful biological sex comparisons. This study used lidocaine as the local anesthetic. Compared to other anesthetics, lidocaine is faster acting with a shorter length of duration³, but it is still estimated to last 0.5-2 hours, which is longer than our study duration.¹⁸ It is possible that the response to other local anesthetics might be different. Future studies could investigate how EDS patients are affected by different types of local anesthetics. This study used a subcutaneous injection into the forearm as a proxy for other medical procedures with local anesthetic. This is not a legitimate clinical context for a medical procedure that a patient would undergo. Further studies may want to explore this resistance in a more clinically reflective context to determine both severity and possible solutions.

Conclusions:

This study provides objective evidence that patients with EDS do experience a shorter duration of local anesthetic effect when using lidocaine, confirming the common anecdotal reports by patients with EDS. This study can help inform clinical choices in anesthesia for both health care providers and patients with EDS.

Key Messages:

What is already known on this topic:

Patients with Ehlers-Danlos syndrome (EDS) frequently report inadequate anesthesia during procedures requiring local anesthetic.

What this study adds:

This clinical trial provides objective evidence that patients with EDS have reduced effectiveness of lidocaine local anesthetic compared to healthy controls.

How this study might affect research, practice, or policy:

This research will help clinicians to recognize potential differences in anesthetic effectiveness in patients with EDS, leading to a better quality of care for this patient group.

Abbreviations:

EDS

Ehlers-Danlos Syndrome

HC

Healthy Controls

5M

5 minutes

15M

15 minutes

30M

30 minutes

ANCOVA

Analysis of covariance

a.u.

Arbitrary units