Article Text

other Versions


MRI evaluation of metal acupuncture needles
  1. Ling Mei1,2,
  2. Xiaojing Long1,
  3. Yanjun Diao1,3,
  4. Haibo Yu3,
  5. Wanzhang Yang4,
  6. Leanna J Standish5,6,
  7. Bensheng Qiu1,5
  1. 1Paul C Lauterbur Research Centers for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong, China
  2. 2School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China
  3. 3Shenzhen International TCM Training Center, Shenzhen Hospital of Traditional Chinese Medicine, Shenzhen, Guangdong, China
  4. 4Department of Rehabilitation, Shenzhen Nanshan Hospital, Shenzhen, Guangdong, China
  5. 5Bastyr University Research Institute, Bastyr University, Kenmore, Washington, USA
  6. 6Department of Radiology, University of Washington School of Medicine, Seattle, Washington, USA
  1. Correspondence to Professor Bensheng Qiu, and Dr Xiaojing Long, Paul C Lauterbur Research Centers for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, 1068 Xueyuan Boulevard, University Town of Shenzhen, Xili Nanshan, Shenzhen, Guangdong 518055, China; bqiu{at} and xj.long{at}


Objective To evaluate the MR compatibility of three metal acupuncture needles (a standard stainless steel needle, a gold needle and an austenitic stainless steel needle) by comparing their imaging artefacts, radiofrequency heating effects and ease of operation.

Methods The MRI artefacts of the three metal needles were first evaluated by placing them in an agar gel phantom and performing MRI of the phantom. The increase in temperature during MRI was recorded using an MR-compatible fibreoptic thermometer. MRI of acupuncture at SP6 was performed using the MR-compatible gold needle and the austenitic stainless steel needle.

Results The standard stainless steel acupuncture needle produced large imaging artefacts on MRI. The gold needle was superior for MRI but not rigid enough for some clinical applications such as scalp acupuncture. The austenitic stainless steel needle is non-ferromagnetic and compatible with MRI. None of these acupuncture needles introduced radiofrequency heating during MRI.

Conclusions The evaluation of MR compatibility showed that gold and austenitic stainless steel needles are MR-compatible and therefore can be used for MRI of acupuncture.

Statistics from


Acupuncture, an important component of Traditional Chinese Medicine, has gradually been recognised and accepted by the Western medical community. However, the underlying mechanisms of acupuncture have not yet been fully established. With the development of functional MRI (fMRI), investigators have begun to explore the correlation between acupuncture points and specific functional areas of the brain.1–7 Traditional metal needles made of ferromagnetic materials8 ,9 are not MR-compatible and generate large artefacts on MRI. Because of the lack of MR-compatible acupuncture needles, researchers have been unable to study the effects of stimulation acupuncture points on the upper torso or the head, such as GV24 Shenting, LI20 Yingxiang and GV14 Dazhui, or scalp acupuncture points used for stroke treatment using MRI. Ferromagnetic needles experience magnetic force within the MRI scanner which may lead to needles being pulled out of tissues and pose a risk to the subject, especially to the head or eyes. In addition, metal acupuncture needles may act as electrical conductors and induce a temperature increase during radiofrequency (RF) transmission in MRI.10 In order for MRI technology to be applied to acupuncture research, needles must be non-magnetic and non-susceptible to RF-related heating that could damage tissues adjacent to the inserted needle.

Beissner et al11 recently evaluated several acupuncture needles and concluded that conventional acupuncture needles should not be used in the MRI environment and suggested that non-ferromagnetic or plastic needles may be used as alternatives. However, the plastic needle they assessed has some disadvantages in comparison with metal needles—for example, it cannot be used for electroacupuncture.

In this study we performed experiments to evaluate MR compatibility of three metal acupuncture needles, including a standard stainless steel needle, a gold needle and an austenitic stainless steel needle, by comparing their imaging artefacts, RF heating effects, cost and ease of operation. In vivo MRI of acupuncture at the lower leg acupuncture point SP6 Sanyinjiao was performed using MR-compatible gold and austenitic stainless steel needles.

Materials and methods

Metal acupuncture needles

The metal acupuncture needles used in this study were a traditional stainless steel acupuncture needle, a 75% pure gold needle and an austenitic stainless steel needle customised for MRI of acupuncture. Details of these three needles are summarised in table 1.

Table 1

Technical specifications of the three acupuncture needles used in the study

The length and diameter of the standard stainless steel needle and the gold needle were 4 cm and 0.3 mm, respectively, whereas the austenitic stainless steel needle was slightly longer and thicker (5 cm long with a diameter of 0.32 mm).

In vitro MRI evaluation of metal acupuncture needles

Experiments were performed to compare MRI artefacts of the needles and to measure temperature increases in an agar gel phantom during MRI scanning. MRI scans were obtained on a whole body 3 T MRI scanner (MAGNETOM Trio Tim, Siemens, Erlangen, Germany) equipped with a body coil for RF transmission.

The needle was inserted into the phantom and a gradient-echo fast low angle shot (FLASH) sequence12 ,13 was used to obtain T1-weighted MRI with TR 100 ms, TE 2.86 ms, field of view (FOV) 150 mm×117 mm, flip angle 70°, matrix size 256×320, slice thickness 5 mm and in-plane resolution 0.6 mm×0.5 mm. Five slices around the needle were obtained to evaluate the imaging artefacts of the needles.

To measure the temperature variation during MRI scanning, the needle was removed from the phantom and attached to an MR-compatible fibreoptic thermometer with a sensitivity of 0.01°C. The tip of the thermometer, where the sensor is situated, was attached to the needle and the attached needle and thermometer were inserted into the phantom. The temperature was recorded every 5 s during MRI scanning. The MRI pulse sequences included: (1) a T2*-weighted echo planar imaging (EPI) sequence14–16 with TR 2000 ms, TE 30 ms, FOV 150 mm×150 mm, flip angle 90° and slice thickness 3 mm; and (2) a T1-weighted gradient-echo FLASH sequence with TR 8 ms, TE 2.86 ms, FOV 150 mm×117 mm, flip angle 70°, matrix size 256×320 and slice thickness 3 mm, which can produce a high specific absorption rate (SAR) to induce a potential heating effect. In both MRI sequences the number of measurements was set to 700 and the total scanning time was 23 min and 32 s (700 TRs). The temperature records were processed using MATLAB 2010 (Matlab, Asheboro, North Carolina, USA), in which the temperatures at every 12 consecutive time points were averaged; 23 average temperatures were obtained and presented as a curve of temperature versus imaging time.

In vivo MRI of acupuncture using metal needles

After MR evaluation of the compatibility of the three needles in the phantom, the gold needle and the austenitic stainless steel needle were used for in vivo MRI of human acupuncture; the standard stainless needle was not used because of the in vitro imaging artefact (see below). SP6, located on the medial side of the leg 10 cm above the tip of the medial malleolus and posterior to the medial border of the tibia, was chosen for needle insertion by a professional acupuncturist. Two orientations, the coronal and transversal planes, were scanned using a fast spin echo sequence with TR 700 ms, TE 27 ms, flip angle 120°, matrix size 512×435 and slice thickness 2 mm.17 ,18 The FOV of the coronal plane was 300 mm×201 mm, while the FOV of the transversal plane was 300 mm×120 mm. After MRI of acupuncture, the needle was removed from the acupuncture point.


In vitro phantom experiment

The T1-weighted MRI of acupuncture needles in the phantom are shown in figure 1. The MRI results show that the standard stainless steel needle produced a large artefact, where most signal generated by the needle and its surrounding was missed. The gold needle generated a relatively small artefact, and the austenitic stainless steel needle also induced relatively small artefact on the images, which provide clear data at most regions within the FOV.

Figure 1

MRI of an agar gel medium with different acupuncture needles: (A) standard stainless steel needle produced a large imaging artefact; (B) gold needle generated a small artefact; and (C) austenitic stainless steel needle produced some artefact on the MRI.

The temperature increases of the three needles during 700-TR MR scanning are shown in figures 2 and 3. For the EPI sequence, all of the temperature curves were almost horizontal, indicating no RF heating risk when using these metal needles (figure 2). For the FLASH sequence, the slope of the increase in temperature was slightly larger than for EPI scanning (figure 3), but the temperature variation was still small and should be safe for patients. The temperature evaluation of the three needles was repeated at least four times.

Figure 2

Temperature changes of the acupuncture needles induced by an echo planar imaging (EPI) sequence. All three needles showed no obvious temperature increase during MRI using an EPI sequence.

Figure 3

Temperature changes of acupuncture needles scanned using a fast low angle shot (FLASH) sequence. Although the FLASH sequence had a high specific absorption rate, the temperature changes of the three needles remained very small.

Comparing the temperature changes of the three metal needles, we found that all the temperatures tended to increase, but the maximum rise was less than 0.5°C and should not be perceptible to the patient nor influence the MRI.

In vivo MRI of acupuncture

The coronal and transversal planes of the in vivo MRI are shown in figure 4. The MRI showed that the gold needle (figure 4A) produced a smaller artefact than the austenitic stainless steel needle (figure 4B). However, it was difficult to determine the exact depth of insertion of the gold needle, especially in muscle, since the contrast between the needle and its surrounding is low. In addition, the gold needle is less rigid than the stainless steel needle and bends during insertion under the skin, particularly in areas of resistance such as the scalp.

Figure 4

MRI of the needles inserted into SP6 (top row: coronal plane, bottom row: transversal plane). (A) The gold needle produced a small artefact; however, the contrast between the needle and its surrounding muscle was low. (B) The austenitic stainless steel needle generated some artefact but it was easy to identify the needle and measure the insertion depth.


Acupuncture needles evolved originally from bian-stone, bamboo and bone to metals such as bronze, gold, silver and stainless steel. However, the same acupuncture manipulation with different kinds of needles may produce different effects. Among these materials, the stiffness and toughness of copper, gold and silver needles is less than that of stainless steel needles. As fMRI is being increasingly used for acupuncture studies, MRI-compatible metal needles have attracted increased attention from investigators. Plastic needles can be used for MRI of acupuncture,11 but they are not commonly used in the clinic because of their low elasticity and extremely low electrical conductivity. Standard stainless steel needles are widely used in the clinic; however, they generate serious artefacts and may experience a magnetic force inside the MRI scanner. Evidence from previous work11 and our present study confirms that standard stainless steel acupuncture needles are unsafe and generate a significant artefact, so cannot be used in MRI scanners. Our experimental results show that the gold needle is superior because of the small imaging artefacts and the high level of safety in MRI. However, it is soft and tends to bend during acupuncture, which makes it difficult to control the insertion depth when being used for certain treatments. In addition, the higher cost makes it less practical for widespread use in the clinic.

The austenitic stainless steel, which differs from the standard stainless steel in its composition and molecular structure,19 is a possible option as an MR-compatible material. The austenitic matrix changes the susceptibility of the normal stainless steel, resulting in a reduction in magnetic field distortion and imaging artefacts.20 In addition, it is stiffer than a gold needle, so it is easier to manufacture and to use for acupuncture studies.

Previous studies have shown that long conductive wires can induce RF heating in the surrounding tissue during MRI.15 ,21 Some investigations showed that the effect may lead to an increase in temperature of more than 60°C at the tip when the resonance conditions were met for an antenna.21 In our study, two commonly used sequences for fMRI of acupuncture, the EPI and FLASH sequences, were used for temperature measurement. EPI is a fast data collection sequence with multiple echoes in an excitation, and the TR of EPI should be set long enough to meet its acquisition scheme.13 The EPI sequence therefore has a lower SAR value, so the RF heating power could spread out during a long TR. To ensure the safety of the needles with regard to heating, we designed our second test at a critical condition using a FLASH pulse sequence with a short TR and high SAR. Although all the temperature measurements showed a tendency to increase, none of the three needles produced an increment larger than 0.5°C during 20 min MR scanning, indicating that none of these metal needles induced RF heating problems during MR scanning. However, it is necessary to conduct further tests to evaluate the safety of MRI when the needles are connected to a long cable for electrostimulation.

In conclusion, we performed a series of in vitro and in vivo experiments to evaluate MR compatibility of three metal acupuncture needles made of standard stainless steel, gold and austenitic stainless steel, respectively. The experimental results showed that the traditional stainless steel needle is not MR-compatible and thus is not recommended for use in MRI. The gold needle and the austenitic stainless steel needle demonstrated higher MR compatibility and smaller imaging artefacts. None of the three needles showed RF heating risk during MRI of acupuncture. The clinical use of gold needles is not practical due to their high cost and softness. The austenitic needle is inexpensive, strong enough to penetrate thick skin such as the scalp and appropriate for electrical stimulation, so it is the most practical type of acupuncture needle for MRI of acupuncture.

Summary points

  • We compared the performance of standard, gold and austenitic stainless steel needles in MRI.

  • The temperature increase was <0.5°C for all three needles.

  • Standard needles gave an obscure image and so were unsuitable for MR; the other two types are MR-compatible and austenitic stainless has some advantages over gold.


View Abstract


  • Contributors BQ and LM designed the experiment. LM performed the experiment, assisted by YD and HY who were licensed acupuncturists. The austenitic stainless steel needle was provided by WY. After the experiment, LM and XL processed the data and wrote the paper which was finally revised by XL, BQ and LJS.

  • Funding This work was supported by National Basic Research Program of China (973 Program)-2010CB732604, Guangdong Innovation Research Team Fund for Low-Cost Healthcare Technologies and the Bairen Talent Project of CAS.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The experiment was approved by the Institutional Review Board of Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.