Mode of action of cupping—Local metabolism and pain thresholds in neck pain patients and healthy subjects

Complementary Therapies in Medicine (2014) 22, 148—158

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevierhealth.com/journals/ctim

Mode of action of cupping—–Local metabolism and pain thresholds in neck pain patients and healthy subjects M. Emerich a, M. Braeunig b, H.W. Clement c, R. Lüdtke d, R. Huber a,∗ a

Center for Complementary Medicine, Department of Environmental Health Sciences, University Medical Center, 79106 Freiburg, Germany b Department Psychosomatic Medicine, University Medical Center, 79106 Freiburg, Germany c Department Child and Youth Psychiatry, University Medical Center, 79106 Freiburg, Germany d Karl und Veronica Carstens-Foundation, 45276 Essen, Germany Available online 7 January 2014

KEYWORDS Lactate; Adenosin; Algometry; Microdialysis; Subcutaneous tissue; Ultrasound

Summary Objectives: Cupping worldwide has been part of traditional medicine systems and is in the western world used as CAM therapy mainly for treating pain syndromes. The mode of action is up to now unclear. In order to investigate its mechanism we measured in parallel metabolic changes in the tissue under the cupping glass and pressure pain thresholds. Design and interventions: In 12 volunteers (6 healthy subjects and 6 patients with chronic neck pain) a microdialysis system was implanted subcutaneously on both sides (left and right) above the trapezius muscle. After baseline measures cupping was performed at one randomly selected side (left or right), the other side served as control. Every 20 min during baseline measures and for 280 min after cupping, microdialysis probes for detection of lactate, pyruvate, glucose and glycerin were taken. In addition, pain thresholds were measured before and after cupping with algometry. Results: Cupping resulted in a strong increase of lactate (beginning 160 min after cupping until the end of the measurements) and the lactate/pyruvate ratio, indicating an anaerobe metabolism in the surrounding tissue. Baseline pain thresholds were non-significantly lower in neck pain patients compared to healthy controls and slightly increased immediately after cupping (p < 0.05 compared to baseline close to the area of cupping in healthy subjects and on the foot in neck pain patients). After 280 min no more significant changes of pain thresholds were detected. Conclusions: Cupping induces >280 min lasting anaerobe metabolism in the subcutaneous tissue and increases immediate pressure pain thresholds in some areas. © 2014 Elsevier Ltd. All rights reserved.

∗ Corresponding author at: Center for Complementary Medicine, Department of Environmental Health Sciences, University Medical Center Freiburg, Breisacher Str. 115 B, 79106 Freiburg, Germany. Tel.: +49 761 270 82010; fax: +49 761 270 83230. E-mail address: [email protected] (R. Huber).

0965-2299/$ — see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctim.2013.12.013

Pain thresholds in neck pain patients

Introduction Cupping is part of Arabian, Chinese and European traditional medicine systems and is used for many different indications.1 The principle is a sucking method. The cupping glass is applied to the skin, mostly to parts of the back of the patient. Because of the vacuum, the skin is sucked into the cupping glass, becomes red and warm, and shows, when the vacuum is strong, signs of sub- and/or intracutaneous bleeding (petechiae). Furthermore, moisture is sucked out of the skin and, in case of wet-cupping, blood is collected in the cupping glass. Within western complementary medicine cupping is mainly used to treat pain and we focused, therefore, on this indication and on dry cupping. Regarding pain cupping has been efficacious in controlled clinical studies for treatment of neck pain,2—4 nocturnal brachialgia,5 osteoarthritis6 and rheumatoid arthritis.7 A recent review of 7 randomized controlled studies found cupping to be promising for pain treatment.8 The mode of action of cupping to reduce pain is, however, unclear. Different hypotheses have been reported. ‘‘Metabolic’’ hypothesis assume, that cupping decreases an increased muscle activity which results in pain reduction (1). The hypothesis is based on the fact, that neck pain patients have been demonstrated in some studies to have an increased muscle activity9—11 and an impaired blood flow compared to healthy controls and in comparison between the painful and not painful side of the neck.11,12 Muscle tone and blood flow are inversely correlated in erector spinae muscles.13 Cupping results in visible redness of the skin of the treated area, local vasodilatation and has been reported to improve microcirculation14 which might improve these conditions. Also the glucose metabolism seems to be disturbed in painful muscles of the neck because intramuscular lactate and pyruvate levels have been different to neck muscles of healthy controls.11,15 Glucose metabolites have not yet been investigated before during and after cupping. In this study we therefore investigated the local concentration of glucose metabolites in the area of cupping. ‘‘Neuronal’’ hypothesis assume that cupping influences chronic pain by altering the signal processing at the level of the nociceptor, the spinal cord and also the brain.16,17 It is expected that the activation of nociceptors by cupping and other naturopathic reflex therapies can stimulate A␦ and Cfibers with involvement of the spino-thalamo-cortical pain pathway. Because the peripheral nociceptor is sensitized by metabolic factors like lactate, adenosine triphosphate, cytokines and others,18 metabolic and neuronal hypothesis are linked together. Secondary aim of our study was to study pain thresholds during cupping simultaneously to metabolic changes. A further hypothesis explains clinical effects through placebo mechanisms. In a previous study we investigated the factors affecting the main physical principle of cupping, namely the negative pressure in the cupping glass (volume, opening area of the cupping glass, mode of creating the negative pressure, experience of the caregiver) in order to standardize it.19 In the present study we wanted to test different hypotheses by investigating local tissue metabolism at the site of

149 cupping and pain thresholds with established methods. To detect potential disease related changes of pain thresholds and local tissue metabolism the study was performed in healthy volunteers and neck pain patients.

Patients and methods Study design Individually controlled, randomized, explorative monocenter study.

Healthy volunteers and patients 6 healthy volunteers and 6 otherwise healthy patients with chronic neck pain (male and female, age 18—50 years for both groups) were recruited by public notice in the University Medical Center Freiburg. Chronic functional neck pain had to persist since at least 12 weeks with an intensity of at least 3 points on a rating scale from 0 (no pain) to 10 points (maximum possible pain) at screening. Exclusion criteria for neck pain patients were an organic cause (e.g. osteoporosis, disk protrusion with neurological abnormalities) of the neck pain, excluded by clinical examination and history. Exclusion criteria for both groups were any acute or chronic disease with the exception of allergies, skin disease on the back, heat urticaria, treatment with anticoagulants, history of bleeding disorders, history of alcohol or drug abuse, insufficient ability to understand German, pregnancy or breast feeding or participation in another study throughout the last three months. All participants gave their written consent before inclusion in the study. Healthy volunteers and patients received a compensation of 100 — Euro for participation in the study. The study was approved by the ethical committee of University Medical Center Freiburg, Germany.

Interventions Algometry: Mechanical pain thresholds were measured with an algometer (Somedics, Hörby, Sweden) in N/cm2 according to the standardized protocol of Rolke.20,21 Briefly, on a 1 cm2 area of the left hand, left leg and on the left and right lower back measurements were performed at baseline (before cupping), immediately after cupping and 140 and 280 min after cupping. Each measurement comprised 3 measures. In each measure pressure with the algometer was increased by 50 kPa/s until the subject/patient signalized pain by pressing a button. Mechanical pain thresholds were measured in N/cm2 . The mean of the respective three values was taken for further analysis. Because algometry could have interfered with the microdialysis measures it was not performed in the neck area but on the lower back. Microdialysis: Microdialysis is an established method to measure e.g. metabolic parameters in human tissues. It is based on the principle of dialysis. A microfilament (CMA AB Solna, Sweden), which contains the semipermeable membrane, was inserted in the subcutaneous tissue on both sides (right and left) of the subjects/patients neck with the help of a spliceable CMA 63 microdialysis catheter under sterile conditions. The length of the membrane was 30 mm, the

150 diameter was 0.9 mm, the molecular cutoff was 20.000 Da. The respective tips of the microfilament were placed above the trapezius muscle. The contralateral microfilament was placed symmetrically and had the same distance to the spine and the upper edge of the scapula. The distance from the tip of the catheter to the surface of the skin was measured by ultrasound (see below). The microfilaments were connected with a CMA 106 syringe which contained the CMA perfusion solution and inserted in a CMA 107 microdialysis pump. Flow rate of the pump was 0.5 ␮l/min. The dialysate was collected in CMA microvials. 10 ␮l was obtained each 20 min. After a run in phase of 40 min the first dialysate was rejected. Then, until the end of the study, after 420 min, every 20 min the microvials with the dialysate were changed. The dialysate was kept frozen at −20 ◦ C for further analysis. 5 probes from each side of the neck, taken during 100 min, served as baseline measures. 100 min for baseline were chosen, because we have seen in pre-study experiments that more than 60 min were needed to reach steady state conditions. The dialysate obtained during cupping was rejected because of the irregular flow during this period. After cupping, microdialysis was continued for another 280 min (14 measures). 420 min after implantation the microdialysis filament was removed and the subjects/patients were allowed to go home. Cupping: 140 min after implantation of the microdialysis system on the right and left side above the trapezius muscle, cupping was performed for 15 min above one of the randomly selected sides in healthy volunteers or, in neck pain patients, above the side with the predominant pain. In addition, cupping was performed for 15 min on the contralateral side of the lower back for the investigation of pain thresholds. Cupping was always performed by the same, experienced investigator. The volume of the cupping glass was 168 ml, the opening area 15.7 cm2 . The cupping glass was provided with a stopcock which could be connected to a pressure gauge (GDH 200-12 Vakuummeter, Greisinger Electronics, Germany) in order to measure the pressure in the cupping glass. The negative pressure was obtained by holding the flame of an alcohol soaked swab for 2 s in the opening of the cupping glass and then immediately pressing the cupping glass onto the skin. Fig. 1 shows the areas of cupping, microdialysis and algometry.

Outcome parameters As parameters for microdialysis pyruvate, lactate, glucose, glycerin and adenosine were selected. Pyruvate (88.06 Da) is an intermediate of aerobe and anaerobe glycolysis and a marker of degradation of carbohydrates. Lactate (90.08 Da) results from non-oxidative metabolism of pyruvate. An isolated increase of lactate can be caused by hypoxia or just an increased metabolism. To discriminate these two reasons, the lactate/pyruvate ratio is used. An increase of lactate and the lactate/pyruvate ratio is proving hypoxia, because cells reduce pyruvate to lactate during hypoxia.22 The lactate/pyruvate ratio is, furthermore, established to estimate the tissue oxygen supply in microdialysis studies.23 Glycerin (92.1 Da) is an indicator of lipolytic activity in the subcutaneous tissue.24 It results from hydrolysis of triacylglycerides

M. Emerich et al. and free fatty acids. Lipolysis is among others regulated by the sympathetic nerve system (release of epinephrine) and, therefore, can indicate changes of this system. Glucose (180.06 Da) in the microdialysate is a marker of the energy supply of the cell. From the rest of the re-frozen dialysate, the nucleotide adenosine was analyzed, because of its potential role for treating chronic pain.25 Analysis of the microdialysis parameters was performed with a CMA 600 analyzer, measuring 20 probes in one session. It is based on measuring photometric absorption after enzymatic degradation. The detection threshold is 0.1 mmol/l for glucose and lactate and 0.01 mmol/l for glycerin and pyruvate. Adenosin was analyzed post hoc from the rest of the probes. Because of the low quantities baseline probes 1—3 and 4—5 as well as two post-cupping probes, respectively, were pooled. Ultrasound (Logiq P5, GE Medical Systems, 7.5 MHz probe) was used to determine the depth of the catheter tip in the tissue. It was measured in millimeter distance from the surface of the skin. Furthermore, ultrasound was used in a single subject to measure the thickness of the subcutaneous tissue before and after cupping. The 6 neck pain patients rated the severity of their neck pain with the standardized neck pain and disability scale (NPDS) in the morning of the day when the investigations were performed and one week later. The NPDS is a 20-item questionnaire that was specifically developed for patients with neck pain and has been proven to be sensitive, reliable and valid for measuring neck pain.26,27 It distinguishes 6 grades of severity: 0—22 points = no to minimal, 23—40 points = mild, 41—57 points = moderate, 58—74 points = moderate to severe, 75—92 points = severe, 93—100 points = extremely severe. All subjects/patients were contacted by telephone the day after the investigation and were asked about side effects. Neck pain patients additionally were contacted one week after the experiment to fill in the NPDS.

Randomization and blinding In healthy subjects the side of the trapezius muscle (right or left) to be cupped was selected by unstratified randomization according to a randomization list. Randomization was concealed (drawing of a sealed opaque envelope). Neck pain patients were asked to mark the most painful area on their neck and this area was chosen for cupping.

Planning of sample size and statistics For sample size calculations we assumed the mean difference of one of the microdialysis parameters between the cupped and the control body side to be 1.0 standard deviation, irrespective whether healthy or neck pain patients were studied. If so n = 12 are needed to detect this effect with a one-sample t-test and a statistical power of ß = 90%.28 Courses of outcome parameters were analyzed by generalized linear models for repeated measurements.29 These models included patient identification, time, and status of patient (healthy or diseased) as independent factors and assumed the within subject correlation to be autoregressive of first order. Treatment effects were estimated via

Pain thresholds in neck pain patients

151

Figure 1 Sites of interventions and measurements on the back. (A) On each side (left and right) of the neck microdialysis filaments (dotted line) were inserted. Above one randomly selected side cupping (circle) was performed. (B) Area on the lower back where pain thresholds were measured (dot). Cupping was performed on the contralateral side (dot with circle).

general estimation equations (GEE) by adequate two-sided t-tests. Appropriate outcome parameters of microdialysis were correlated with the intervention. For each parameter p-values were multiply adjusted by the Bonferoni—Holm30 procedure.

Results All 12 subjects and patients finished the study per protocol. Except for the typical petechiae no side effects from cupping or the microdialysis at the end of the experiment or at follow up one day later were reported. Characteristics of the participants are shown in Table 1. There were no differences between the groups regarding age, body mass index (BMI), negative pressure in the cupping glass and depth of the catheter in the tissue. Single values of microdialysis parameters had to be excluded from analysis because of pump dysfunction or trapped air in the CME analyzer. The NPDS was reduced from 33.2 ± 13.5 before cupping to 27.7 ± 11.4 one week later. The difference was not statistically significant.

Changes observed irrespective of groups Algometry: The course of the pain thresholds after cupping was heterogeneous. Analyzing the whole collective, no significant changes in comparison to baseline could be found (data not shown) Microdialysis: Fig. 2 gives an overview of the course of pyruvate, lactate, glucose and glycerin in the cupping area of the whole collective in comparison to baseline. It shows a strong increase of lactate about four standard deviations with a plateau after 200 min which is accompanied by an increase of pyruvate at half strength. Glycerin falls about two standard deviations and slowly returns. Glucose shows a short rise and falls back to normal. Table 2 shows the mean differences between the area of cupping and the control area for pyruvate, lactate and glycerin and the results of the GEE-ANCOVA analysis. For glucose, except at a single time point 60 min after cupping (increase 0.6 mmol in the cupping area, p = 0.002 adjusted to Bonferoni—Holm) no significant changes and no time periods with a trend could be found. Lactate significantly and consistently increased also in comparison to the control area, with a maximum

152 Table 1

M. Emerich et al. Characteristics of the study participants. Healthy volunteers

Neck pain patients

Whole collective

N Age (years) BMI kg/m2 Gender (female/male) Negative pressure in the cupping glass (hPa)

6 24.7 ± 1.0 21.6 ± 3.0 4/2 315 ± 64

6 25.2 ± 1.3 22.2 ± 1.7 4/2 283 ± 54

12 24.9 ± 1.2 21.9 ± 2.4 8/4 299 ± 59

Depth of the catheter in the tissue (mm) Side of cupping Control side

0.54 ± 0.11 0.53 ± 0.10

0.53 ± 0.14 0.57 ± 0.07

0.53 ± 0.12 0.55 ± 0.08

Pyruvate (␮mol/l) mean baseline Side of cupping Control side

86 ± 57 126 ± 50

82 ± 27 95 ± 50

84 ± 42 111 ± 50

Lactate (mmol/l) mean baseline Side of cupping Control side

0.8 ± 0.4 1.3 ± 0.8

0.9 ± 0.3 1.6 ± 0.9

0.9 ± 0.3 1.4 ± 0.8

Glycerin (␮mol/l) mean baseline Side of cupping Control side

131 ± 118 180 ± 137

115 ± 42 92 ± 33

123 ± 85 136 ± 106

Glucose (mmol/l) mean baseline Side of cupping Control side

2.9 ± 1.3 2.5 ± 0.8

2.2 ± 0.6 3.0 ± 0.7

2.6 ± 1.1 2.7 ± 0.8

220 min after cupping. 280 min after cupping values had not returned to the level of the control area. Pyruvate slightly and less consistently increased in comparison to the control area, with a maximum 140 min after cupping. It returned to the level of the control area after about 240 min. Only two (after 140 and 180 min, see Table 2) adjusted p-values but the most non adjusted p-values were significantly different

between 80 and 220 min after cupping (80 min, p = 0.041, 100 min, p = 0.033, 120 min, p = 0.121, 140 min, p < 0.001, 160 min, p = 0.127, 180 min, p = 0.003, 200 min, p = 0.040, 220 min, p = 0.049, all others p > 0.1). Glycerin showed an initial strongly significant decrease 40—80 min after cupping compared to the control side. As a trend levels remained below the control side until the end of the experiment.

Figure 2 Time course of lactate, pyruvate, glucose and glycerin (changes in standard deviations and smoothed curves) of 6 healthy volunteers and 6 neck pain patients (n = 12) after cupping. Baseline is taken from the first five probes before cupping at t = 0. After replacing outliers by medians the individual signals are averaged to the mean. Thick lines are local polynomial fits.

Time after baseline (min)

20 40 60 80 100 120 140 160 180 200 220 240 260 280

Pyruvate (␮mol/l)

Lactate (mmol/l)

Glycerin (␮mol/l)

Mean difference and interval of confidence

p-Value

Mean difference and interval of confidence

p-Value

Mean difference and interval of confidence

p-Value

16 (-5 to 37) 6(-13 to 25) 8 (-8 to 24) 25 (1—49) 18 (1—34) 26 (-7 to 58) 30 (13—47) 17 (-5 to 40) 27 (9—44) 27 (1—52) 34 (0—68) 19 (-7 to 45) 18 (-13 to 50) 29 (-9 to 66)

0.970 0.970 0.970 0.436 0.396 0.970 0.007 0.970 0.036 0.436 0.437 0.970 0.970 0.970

0.1 0.0 0.3 0.6 0.0 0.5 0.5 0.7 0.7 0.9 1.4 1.1 1.2 1.2

1.000 1.000 0.235 0.068 1.000 0.700 0.080 0.003 0.180 0.026 0.001 0.001 0.003 0.044

−10 (-28 to 8) −51.6 (-84 to -19) −72 (-114 to -30) −46 (-72 to -21) −50 (-85 to -14) −42 (-93 to 9) −37 (-71 to -3) −42 (-76 to -7) −40 (-73 to -7) −47 (-83 to -10) −37 (-64 to -10) −38 (-68 to -8) −38 (-68 to -8) −23 (-52 to 5)

0.319 0.023 0.009 0.005 0.067 0.319 0.126 0.113 0.113 0.113 0.079 0.113 0.113 0.319

(-0.1 to 0.3) (-0.3 to 0.3) (0.0—0.6) (0.1—1.0) (-0.6 to 0.6) (-0.2 to 1.2) (0.1—0.9) (0.3—1.0) (0.1—1.3) (0.3—1.5) (0.7—2.0) (0.5—1.6) (0.6—1.9) (0.4—2.0)

Pain thresholds in neck pain patients

Table 2 Differences between cupping area and control area for pyruvate, lactate and glycerin-levels in subcutaneous tissue. Positive values indicate an increase in the cupping area. Mean changes, interval of confidence and p-values adjusted to Bonferoni—Holm are presented.

153

154

M. Emerich et al. 30 control side cupping

lactate/pyruvat ratiio

25

20

15

10

5

0 -100 -80

-60

-40

-20

20

40

60

80

100 120 140 160 180 200 220 240 260 280

minutes

Figure 3

Lactate/pyruvate ratio of medians of area of cupping and control side (n = 12).

The lactate/pyruvate ratio increased in the area of cupping (Fig. 3). Adenosin did not significantly change during the course of the study. There were no differences compared to baseline or between the area of cupping and the control side (data not shown). Maximum lactate concentrations correlated to the extent of the vacuum in the cupping glass (r = 0.81, p = 0.049) in healthy subjects. On the control side no correlation was found. Glycerin levels were, as a trend, negatively correlated to the extent of the vacuum in the cupping glass (r = −0.41, p = 0.180). Ultrasound: Cupping resulted in an edema of the subcutaneous tissue, which was documented by ultrasound in one subject. The thickness of the subcutaneous tissue increased from about 5 mm before to 9 mm immediately after cupping. 130 min after cupping it was 7.5 mm, 13 h later it was reduced to 5.6 mm (Fig. 4A—D).

Comparison between healthy controls and neck pain patients Algometry: Baseline and course of the mechanical pain thresholds after cupping are shown in Table 3. In all tested areas healthy subjects had at baseline higher pain thresholds than neck pain patients. The difference was, however, not statistically significant (p > 0.05). Immediately after cupping pain thresholds tended to increase in all tested areas of neck pain patients (Table 3). A significant increase was found on the foot (p = 0.019 adjusted to Bonferoni—Holm). Healthy volunteers had only in the control area higher pain thresholds in comparison to baseline (p = 0.02 adjusted to Bonferoni—Holm). 140 min after cupping the pain threshold on the foot in neck pain patients was lower than before cupping (p = 0.019 adjusted to Bonferoni—Holm). Significant differences between healthy subjects and neck pain patients could be found on the foot immediately after cupping (p = 0.01 adjusted to Bonferoni—Holm, Table 3). Microdialysis: Comparing healthy volunteers and neck pain patients baseline mean values for pyruvate, lactate, glucose and glycerin were not different (p > 0.1). The course

of the differences between the area of cupping and the control area separated for healthy volunteers and neck pain patients are shown in Fig. 5A—D. For pyruvate and lactate there were no significant cupping induced differences between healthy volunteers and neck pain patients after adjustment for multiple testing. Glycerin levels increased on the control side in healthy volunteers after cupping (Fig. 5C). Except 100 min after cupping (p = 0.031) the differences were, however, not significant when adjusted for multiple testing. Also for glucose differences between healthy volunteers and neck pain patients could be found (Fig. 5D). Healthy volunteers had non-significantly lower values in the control area and non-significantly higher values in the cupping area without changes over time (data not shown). This accumulated to in part (after 100, 180 and 220 min, adjusted p = 0.012, 0.015 and 0.001, respectively) significantly higher differences between control area and area of cupping in healthy volunteers. Neck pain patients had a slightly higher and earlier increase of the lactate/pyruvate ratio after cupping, compared to healthy volunteers (Fig. 6). The adenosin concentration did not show a group difference between neck pain patients and healthy volunteers. Baseline mean lactate values tended in neck pain patients to correlate with the depth of the microdialysis system under the skin (r = 0.68 and r = 0.67 in the area of cupping and control side, respectively). In healthy subjects, there was no correlation (r = 0.14 and 0.00, respectively).

Discussion In our study we investigated effects of cupping on mechanical pain thresholds and subcutaneous metabolic parameters in order to test hypothesis about the mode of action of cupping. For measuring metabolic parameters we chose an individually controlled parallel design, comparing the cupped and un-cupped side. We found, that cupping significantly and long lasting increases lactate levels in the subcutaneous tissue. The increased lactate/pyruvate ratio proofs that the lactate

Pain thresholds in neck pain patients

155

Figure 4 (A—D) Thickness of the subcutaneous tissue above the trapezius muscle measured by ultrasound before and after cupping with a negative pressure of 200 hPa. (A) Before cupping (0.47 cm), (B) immediately after cupping (0.89 cm), (C) 130 min after cupping (0.75 cm) and (D) 18 h after cupping (0.56 cm).

resulted from hypoxia. The source of lactate can be from the subcutaneous tissue itself, destroyed erythrocytes or the underlying muscle, which is connected by blood flow with the subcutaneous tissue.31 A mathematical model assuming similar conditions (diameter of the cupping glass 5 cm, pressure −300 hPa, thickness of the subcutaneous tissue 1 cm) as in our study showed, that the pressure on the underlying muscle is still between −250 hPa (upper part) and −180 hPa (lower part, about 3 cm from the skin), corresponding to 187.5 and 135 mmHg, respectively.32 This means, that the blood flow in normotensive subjects (systolic arterial pressure 120 mmHg) is

even in the underlying muscle completely suppressed during cupping. Muscle tissue has, according to several microdialysis studies33—35 at rest 2—3.5 times higher lactate levels than subcutaneous tissue. The subcutaneous tissue itself has only limited capacity to produce lactate. A recent C13 labeled-glucose based microdialysis study showed that only about 25% of subcutaneous tissue lactate originates from the subcutaneous tissue itself, the rest comes via blood flow or diffusion from the muscle below.36 This indicates that lactate originated in our study to a major part from outside the subcutaneous tissue, namely from the underlying trapezius muscle and/or from destroyed erythrocytes.

Table 3 Mechanical pain thresholds measured at different sites. Means and standard deviations of the group of healthy volunteers and the group of neck pain patients (each n = 6) are presented. Baseline (N/cm2 )

Immediately after cupping (N/cm2 )

140 min after cupping (N/cm2 )

280 min after cupping (N/cm2 )

Healthy volunteers Area of cupping (lower back) Control area (lower back) Hand Foot

332 331 185 317

± ± ± ±

158 180 23 104

333 370 169 311

± ± ± ±

181 204* 37 96

321 315 188 326

± ± ± ±

125 156 14 117

351 372 200 291

± ± ± ±

179 208 27 68

Neck pain patients Area of cupping (lower back) Control area (lower back) Hand Foot

280 288 168 252

± ± ± ±

93 109 41 74

302 300 185 283

± ± ± ±

55 130 61 87*

308 317 161 234

± ± ± ±

63 88 53 78*

293 304 180 259

± ± ± ±

80 95 61 102

*

p adjusted to Bonferoni—Holm <0.05 compared to baseline.

156

M. Emerich et al.

A 3,5

B Healthy

Pyruvate side difference μmol/l

Lactate side difference mmol/l

3

Neck pain

2,5 2 1,5 1 0,5 0 -0,5 -1 -1,5 -100

-50

0

50

100

150

200

250

Healthy

100,00

Neck pain

50,00

0,00

-50,00

-100,00 -100

300

-50

0

Time aer cupping [min]

C

150

D

Glycerine side difference mmol/l

100

Glucose side difeerence μmol/l

50

0 -50 -100 -150 -200

Healthy

-250

Neck pain -50

0

50

100

100

150

200

250

300

150

200

250

3

300

Healthy

2

Neck pain

1 0 -1 -2 -3 -100

-300 -100

50

Time aer cupping [min]

-50

0

50

100

150

200

250

300

Time aer cupping [min]

Time aer cupping [min]

Figure 5 (A—D) Means and standard deviation of subcutaneous lactate (A), pyruvate (B), glycerin (C) and glucose (D) in neck pain patients (n = 6) and healthy controls (n = 6) before and up to 280 min after cupping. Side differences between cupping area and control area (contralateral side of the neck) are presented. Higher values mean an increase compared to the control area.

The baseline lactate levels in our study are comparable to other microdialysis studies in subcutaneous tissue reporting 0.8—1.8 mmol/l.36—38 Lactate results, as well as hypoxia/hyperkapnia per se, in local acidosis.39 Lactate and acidosis have for a long time been regarded as responsible for organ damage. Better controlled studies since the 1990s showed, however, that acidosis in skeleton muscle can prolong time

to exhaustion and is protective against potassium induced depression of muscle contraction during exercise.40,41 Furthermore, moderate acute local metabolic acidosis results in local vasodilatation and increased blood flow42—45 by direct pH effects42 and augmenting nitric oxide release in vitro and in vivo.46,47 Neck pain patients tend to have a decreased blood flow and elevated muscle lactate levels in the trapezius muscle compared to healthy controls.11,12,48 Cupping

35 healthy

lactate/pyruvat ratiio

30

neck pain patients

25 20 15 10 5 0 -100 -80

-60

-40

-20

20

40

60

80

100 120 140 160 180 200 220 240 260 280

minutes

Figure 6

Lactate/pyruvate ratio of medians of neck pain patients and healthy controls in the area of cupping (n = 6).

Pain thresholds in neck pain patients per se aggravates these features. After cupping, however, vasodilatation can be observed and improved microcirculation has been reported.13 This might, after a few days, result in a reset to normal of the chronically disturbed interplay between nerval afferences, efferences and local tissue metabolism in neck pain patients. Future studies should directly measure muscle blood flow and muscle activity before and after cupping. In a controlled study3 a single cupping treatment resulted in significantly less pain compared to controls 3 days later. In our study NPDS scores decreased only slightly one week after cupping. Our neck pain patients had no elevated tissue lactate or pyruvate levels compared to healthy volunteers. These differences to other studies may be explained by the only mild symptoms and relatively short duration (mean 19 months) of our neck pain patients. The lactate/pyruvate ratio, however, increased slightly more and earlier compared to healthy controls (Table 1 and Fig. 5). The decrease of glycerin in our investigation can be explained by anti-lipolytic effects of acidosis. Acidosis reduces the norepinephrine induced lipolysis.49 Furthermore, the subcutaneous edema (Fig. 3) might have contributed to the decrease of glycerin. Glycerin release of muscle tissue is only 20—25% of subcutaneous tissue.50 If the edema fluid comes from the muscle it results in dilution and decreased glycerin levels in the subcutaneous tissue. Neuronal hypotheses assume, that cupping modulates functional pain processing in neck pain patients. Indeed, neck pain patients had in another pilot study lower mechanical pain thresholds compared to controls.51 Corresponding, our neck pain patients, had as a trend, lower baseline pain thresholds than healthy controls. Up to now an effect of cupping on pain thresholds has not been clearly demonstrated. In two studies2,3 mechanical pain thresholds were after cupping significantly higher than in waiting list controls but no increase of pain thresholds compared to baseline was reported. In our patients, initially after cupping pain thresholds were at least in part significantly increased (Table 3). This might be explained by counter irritation. 280 min later, however, neither in the area of cupping nor in the control area relevant changes compared to baseline or compared to healthy controls could be found. A limitation is that we could not measure the local cupping effect on the neck, because this might have interfered with the microdialysis measures. Taken together, short term increases of mechanical pain thresholds have been for the first time demonstrated after cupping. They last, however, for less than 280 min and are, therefore, not related to the metabolic changes measured by microdialysis.

Conclusions Cupping induces >280 min lasting anaerobe metabolism in the subcutaneous tissue and increases immediate pressure pain thresholds in some areas.

Conflict of interests The authors have no direct or indirect financial or personal relation with the commercial identities mentioned in this publication.

157

Role of the funding source The study was financed without external funding.

References 1. Abele J. Das Schröpfen-Eine bewährte alternative Heilmethode. München: Urban & Fischer; 2003. 2. Lauche R, Cramer H, Choi KE, Rampp T, Saha FJ, Dobos GJ, et al. The influence of a series of five dry cupping treatments on pain and mechanical thresholds in patients with chronic nonspecific neck pain — a randomised controlled pilot study. BMC Complement Altern Med 2011;11:63. 3. Lauche R, Cramer H, Hohmann C, Choi KE, Rampp T, Saha FJ, et al. The effect of traditional cupping on pain and mechanical thresholds in patients with chronic nonspecific neck pain: a randomised controlled pilot study. Evid Based Complement Alternat Med 2012;2012:429718. 4. Kim TH, Kang JW, Kim KH, Lee M, Kim JE, Kim JH, et al. Cupping for treating neck pain in video display terminal (VDT) users: a randomized controlled pilot trial. J Occup Health 2013;54(6):416—26. 5. Lüdtke R, Albrecht U, Stange R, Uehleke B. Brachialgia paraesthetica nocturna can be relieved by wet cupping — results of a randomised pilot study. Complement Ther Med 2006;14(4):247—53. 6. Teut M, Kaiser S, Ortiz M, Roll S, Binting S, Willich SN, et al. Pulsatile dry cupping in patients with osteoarthritis of the knee — a randomized controlled exploratory trial. BMC Complement Altern Med 2012;12(October):184, http://dx.doi.org/10.1186/1472-6882-12-184. 7. Ahmed SM, Madbouly NH, Maklad SS, Abu-Shady EA. Immunomodulatory effects of blood letting cupping therapy in patients with rheumatoid arthritis. Egypt J Immunol 2005;12(2):39—51. 8. Kim JI, Lee MS, Lee DH, Boddy K, Ernst E. Cupping for treating pain: a systematic review. Evid Based Complement Alternat Med 2011;2011:467014. 9. Juul-Kristensen B, Clausen B, Ris I, Jensen RV, Steffensen RF, Chreiteh SS, et al. Increased neck muscle activity and impaired balance among females with whiplash-related chronic neck pain: a cross-sectional study. J Rehabil Med 2013;45(4):376—84. 10. Park KN, Kwon OY, Ha SM, Kim SJ, Choi HJ, Weon JH. Comparison of electromyographic activity and range of neck motion in violin students with and without neck pain during playing. Med Probl Perform Art 2012;27(4):188—92. 11. Sjøgaard G, Rosendal L, Kristiansen J, Blangsted AK, Skotte J, Larsson B, et al. Muscle oxygenation and glycolysis in females with trapezius myalgia during stress and repetitive work using microdialysis and NIRS. Eur J Appl Physiol 2010;108(4):657—69. 12. Larsson R, Oberg PA, Larsson SE. Changes of trapezius muscle blood flow and electromyography in chronic neck pain due to trapezius myalgia. Pain 1999;79(1):45—50. 13. Masuda T, Miyamoto K, Shimizu K. Intramuscular hemodynamics in bilateral erector spinae muscles in symmetrical and asymmetrical postures with and without loading. Clin Biomech (Bristol, Avon) 2006;21(3):245—53. 14. Cui S, Cui J. Progress of researches on the mechanism of cupping therapy. Zhen Ci Yan Jiu 2012;37(6):506—10. 15. Gerdle B, Lemming D, Kristiansen J, Larsson B, Peolsson M, Rosendal L. Biochemical alterations in the trapezius muscle of patients with chronic whiplash associated disorders (WAD) — a microdialysis study. Eur J Pain 2008;12(1):82—93. 16. Musial F, Michalsen A, Dobos G. Functional chronic pain syndromes and naturopathic treatments: neurobiological foundations. Forsch Komplementmed 2008;15(2):97—103.

158 17. Musial F, Spohn D, Rolke R. Naturopathic reflex therapies for the treatment of chronic back and neck pain. Part 1. Neurobiological foundations. Forsch Komplementmed 2013;20(3):219—24. 18. Staud R. Peripheral pain mechanisms in chronic widespread pain. Best Pract Res Clin Rheumatol 2011;25(2):155—64. 19. Huber R, Emerich M, Braeunig M. Cupping — is it reproducible? Experiments about factors determining the vacuum. Complement Ther Med 2011;19(2):78—83. 20. Rolke R, Magerl W, Campbell KA, Schalber C, Caspari S, Birklein F, et al. Quantitative sensory testing: a comprehensive protocol for clinical trials. Eur J Pain 2006;10(1):77—88. 21. Rolke R, Baron R, Maier C, Tölle TR, Treede RD, Beyer A, et al. Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): standardized protocol and reference values. Pain 2006;123(3):231—43. 22. Harken AH. Lactic acidosis. Surg Gynecol Obstet 1976;142(4):593—606. 23. Jansson K, Ungerstedt J, Jonsson T, Redler B, Andersson M, Ungerstedt U, et al. Human intraperitoneal microdialysis: increased lactate/pyruvate ratio suggests early visceral ischaemia. A pilot study. Scand J Gastroenterol 2003;38(9):1007—11. 24. Dodt C, Lönnroth P, Wellhöner JP, Fehm HL, Elam M. Sympathetic control of white adipose tissue in lean and obese humans. Acta Physiol Scand 2003;177(3):351—7. 25. Zylka MJ. Pain-relieving prospects for adenosine receptors and ectonucleotidases. Trends Mol Med 2011;17(4):188—96. 26. Goolkasian P, Wheeler AH, Gretz SS. The neck pain and disability scale: test-retest reliability and construct validity. Clin J Pain 2002;18(4):245—50. 27. Blozik E, Kochen MM, Herrmann-Lingen C, Scherer M. Development of a short version of the neck pain and disability scale. Eur J Pain 2010;14(8):864e1—7e. 28. Dupont WD, Plummer WD. Power and sample size calculations for studies involving linear regression. Control Clin Trials 1998;19(6):589—601. 29. Diggle P, Liang K, Zeger S. Analysis of Longitudinal Data. Oxford: Clarendon Press; 1994. 30. Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat 1979;6(2):65—70. 31. Cariou JL. 1984—1994: ten years of skin flaps. Improvements and conceptual developments. Development of vascular concepts, classification and clinical concepts. Ann Chir Plast Esthet 1995;40(5):447—525. 32. Tham LM, Lee HP, Lu C. Cupping: from a biomechanical perspective. J Biomech 2006;39(12):2183—93. 33. Hagström-Toft E, Enoksson S, Moberg E, Bolinder J, Arner P. Absolute concentrations of glycerol and lactate in human skeletal muscle, adipose tissue, and blood. Am J Physiol 1997;273(3 Pt. 1):E584—92. 34. Rosdahl H, Hamrin K, Ungerstedt U, Henriksson J. Metabolite levels in human skeletal muscle and adipose tissue studied with microdialysis at low perfusion flow. Am J Physiol 1998;274(5 Pt. 1):E936—45.

M. Emerich et al. 35. Qvisth V, Hagström-Toft E, Moberg E, Sjöberg S, Bolinder J. Lactate release from adipose tissue and skeletal muscle in vivo: defective insulin regulation in insulin-resistant obese women. Am J Physiol Endocrinol Metab 2007;292(March (3)):E709—14. 36. Gustafsson J, Eriksson J, Marcus C. Glucose metabolism in human adipose tissue studied by 13C-glucose and microdialysis. Scand J Clin Lab Invest 2007;67(2):155—64. 37. Jansson PA, Smith U, Lönnroth P. Evidence for lactate production by human adipose tissue in vivo. Diabetologia 1990;33(4):253—6. 38. Jansson PA, Larsson A, Smith U, Lönnroth P. Lactate release from the subcutaneous tissue in lean and obese men. J Clin Invest 1994;93(1):240—6. 39. Das JB, Joshi ID, Philippart AI. Continuous monitoring of pH in the tissue mode: evaluation of a miniature sensor during acidosis and tissue hypoperfusion. J Pediatr Surg 1983;18(6):914—21. 40. Cairns SP. Lactic acid and exercise performance: culprit or friend? Sports Med 2006;36(4):279—91. 41. Bandschapp O, Soule CL, Iaizzo PA. Lactic acid restores skeletal muscle force in an in vitro fatigue model: are voltagegated chloride channels involved? Am J Physiol Cell Physiol 2012;302(7):C1019—25. 42. Aalkjaer C, Poston L. Effects of pH on vascular tension: which are the important mechanisms? J Vasc Res 1996;33(5):347—59. 43. Aalkjaer C, Peng HL. pH and smooth muscle. Acta Physiol Scand 1997;161(4):557—66. 44. Modin A, Björne H, Herulf M, Alving K, Weitzberg E, Lundberg JO. Nitrite-derived nitric oxide: a possible mediator of ‘acidicmetabolic’ vasodilation. Acta Physiol Scand 2001;171(1):9—16. 45. Wang X, Wu J, Li L, Chen F, Wang R, Jiang C. Hypercapnic acidosis activates KATP channels in vascular smooth muscles. Circ Res 2003;92(11):1225—32. 46. Hattori K, Tsuchida S, Tsukahara H, Mayumi M, Tanaka T, Zhang L, et al. Augmentation of NO-mediated vasodilation in metabolic acidosis. Life Sci 2002;71(12):1439—47. 47. Celotto AC, Restini CB, Capellini VK, Bendhack LM, Evora PR. Acidosis induces relaxation mediated by nitric oxide and potassium channels in rat thoracic aorta. Eur J Pharmacol 2011;656(1-3):88—93. 48. Rosendal L, Larsson B, Kristiansen J, Peolsson M, Søgaard K, Kjaer M, et al. Increase in muscle nociceptive substances and anaerobic metabolism in patients with trapezius myalgia: microdialysis in rest and during exercise. Pain 2004;112(3):324—34. 49. Hjemdahl P. Inhibition of the lipolytic response to nerve stimulation during acidosis. Acta Physiol Scand 1976;98(1):80—4. 50. Bolinder J, Kerckhoffs DA, Moberg E, Hagström-Toft E, Arner P. Rates of skeletal muscle and adipose tissue glycerol release in nonobese and obese subjects. Diabetes 2000;49(5):797—802. 51. Javanshir K, Ortega-Santiago R, Mohseni-Bandpei MA, Miangolarra-Page JC, Fernández-de-Las-Pe˜ nas C. Exploration of somatosensory impairments in subjects with mechanical idiopathic neck pain: a preliminary study. J Manipulative Physiol Ther 2010;33(7):493—9.