Although the efficacy of acupuncture in treatment of various diseases has been empirically established, systematic demonstration of this point has been rarely made. Our animal experiments designed to prove the efficacy of acupuncture on a scientific basis are reviewed here.
Each of the following experiments was carried out on two to five groups of mice. Mice in acupuncture group (A-animals) were acupunctured at four to seventeen points on every second or third day beginning with the first experimental day.
Exp. I. The preventive effect of acupuncture on acute poisoning by carbon tetrachloride-With carbon tetrachloride (4.16g/kg body weight) injected in the abdominal cavity on the 18th exp. day, all the non-acupunctured mice (N-animals) died in 72 hours after the administration, while 16% of the A-animals survived.
Exp. II. The curative effect of acupuncture on subacute liver injury induced by carbon tetrachloride. -Carbon tetrachloride (3.47g/kg b. w.) was administered on the 17th exp. day in the same manner as Exp. I. Levallorphan tartrate (antagonist to narcotics) was given to half of the A-animals (L+A-animals) and control animals (L-animals) before each session of acupuncture. Electron microscopic examination revealed severely injured liver cells in the N-animals and very little changes in the A-animals. The liver of L+A-animals was injured to some extent. Since the explanation of toxicity of levallorphan was refuted by the L-animals, this result suggests that some narcotic substances play a role in the process.
Exp. III. The effect of acupuncture on alloxan diabetes. -Alloxan (6mg) was administered on the 8th exp. day, and the animals were sacrificed on the 30th or 46th exp. day for examination. The N-animals showed extremely high blood sugar, though the A-animals did not. Ultrastructural study revealed that characteristic changes of alloxan diabetes in the pancreas had almost disappeared from the A-animals.
Exp. IV. The effect of acupuncture on renal injury induced by mercuric chloride. -With mercuric chloride (0.2mg/100g b. w.) administered three times a week, 10% of the N-animals and none of the A-animals died during the three-month period. Fairly heavy proteinuria (30-100mg/ml) was detected in the N-animals, though uria of A-animals was almost normal. After a one-month period, the N-animals showed a significant weight loss compared to the A-animals. Through the use of the electron microscope, severe injuries were seen in the epithelium of the renal tubules of the N-animals, and healed injuries in that of the A-animals.
Ultrastructural evidence has been presented that acupuncture is effective in (1) preventing and curing liver injury induced by carbon tetrachloride, (2) curing alloxan diabetes, (3) curing nephritis induced by mercuric chloride. Our hypothesis suggesting the possible mechanism that underlies these effects produced by acupuncture has been briefly explained.

It is well known that acupuncture has a curative effect for some chronic diseases such as neuralgy or lumbago, if it is administered carefully in terms of the patient. However, the mechanism of acupuncture's effectiveness remains obscure.
In this experiment, using mice of the ICR strain, the protective effect of acupuncture against mouse liver injury was investigated following the administration of carbon tetrachloride.
Experiment I (Fig. 1): One hundred mice were divided into two groups. The first group of animals was administered carbon tetrachloride ip on day 18. The second group was treated a total of six times with acupuncture every three days from the first experimental day to day 18 at the points of Tienshu (ST-25), Feishu (BL-13), Hsinshu (BL-15), Kanshu (BL-18), Pishu (BL-20), Weishu (BL-21), Chimen (LV-14), Chihai (CV-6), Chungwan (CV-12) and Chuchueh (CV-14) (Figs. 3, 4). Fllowing the administration of carbon tetrachloride, the mouse mortality was checked. All of the animals of the first group (carbon tetrachloride administration only) died within 72 hours, but only 84% of the animals of the second group (carbon tetrachloride and acupuncture treatments) died (Fig. 2).
Experiment II (Fig. 1): One hundred and 60 mice were divided into 5 groups. The first group was used for the control (Fig. 5). The second group was injected ip with carbon tetrachloride of 3.47g/Kg BW on day 17. The third group was given a total of 5 acupuncture treatments every three days at the points as mentioned in Experiment I, and carbon tetrachloride was injected as in the second group.
The fourth group was injected a total of 5 times with 0.0004mg levallorphan (an anti-opiate) every three days. The fifth group was given acupuncture treatments like the third group; levallorphan was injected before the acupuncture, and also carbon tetrachloride was injected on day 17 as in the second or third, group.
The animals were sacrificed on day 20, and the liver tissues were treated as usual and then observed by electron microscopy.
The hepatocytes of the second group (carbon tetrachloride only) were heavily injured, and contained, a number of fat droplets, autophagic vacuoles and degenerated cell organelles (Figs. 6, 7, 8, 9). On the other hand, the hepatocytes of the third group (carbon tetrachloride and acupuncture treatments) were almost normal, and without any cell injury (Figs. 10, 11, 12).
The hepatocytes of the fourth group were not injured with levallorphan, suggesting that levallorphan has no cell toxicity (Fig. 13).
The fifth group showed hepatic cell injury similar to the second group (Fig. 14). Levallorphan thus might have an antagonistic action for the acupuncture effectiveness.
In summary, it was concluded that acupuncture has a protective effect against liver injury caused by carbon tetrachloride administration, and prevents not only animal death but also hepatic cell injury. The mechanism of the acupuncture effectiveness is speculated to be as follows. The stimulus of the acupuncture is conducted to the central nervous system and then the central nervous system secretes some neural hormonal substances such as endorphin or enkephalin, which might regulate or accelerate the autonomic nervous system, and then the viscerae functions are accelerated and thus might repair the injuries or dieseases.
In the above-mentioned experiments, levellorphan may have disturbed the secretion or action of endorphin or enkephalin from the central vervous system, which was said to be secreted following the acupuncture treatments, resulting in the ineffectiveness of acupuncture treatments.
Fig. 1. Schedules of the experiments.
Fig. 2. The mortality rate of mice following carbon tetrachloride administration (Experiment I).
Fig. 3. Acupuncture points on the mouse belly.
Fig. 4. Acupuncture points on the mouse back.
Fig. 5. A part of the hepatic cell from a normal mouse. The nucleus (N) and nucleolus (NO) are morphologic

The acupuncture points are quite important in Oriental medicine. They have been known for about 3, 000 years. However, the points have not been ascertained morphologically to date.
In this experiment, the acupuncture points such as Hoku (LI-4), Taichong (LV-3) and one of the auricular points, which were at first obtained morphologically using a Rydoraku point-searching apparatus (12V, DC), and then needled perpendicularly with acupuncture needle No. 30, staining the tip of the needle with carbon black ink to mark the point localizations.
Small pieces of skin were marked with ink cuts and fixed with 10% neutralized formalin, and then serial paraffin sections were made for light microscopical observations.
The acupuncture points seemed to be the places at which the electric resistance was usually lower than at other non-acupuncture points.
Non-acupuncture points with high electric resistance were also taken for the control (Fig. 7).
From the light microscopical observations of the acupuncture points, a special complex was found (Figs. 2, 3, 4, 5, 6). It was composed of a nerve fiber running horizontal to the surface of the skin, some blood and lymph vessels and a small amount of collagenous fibers.
These elements seemed to be a plexus forming a complex, and the complex was usually located within the subcutaneous tissue.
On the other hand, such complex had not been observed at the non-acupuncture points so far (Fig. 7).
In summary, it is postulated that the acupuncture points seem to be in some way related to the nervous elements, the vascular system and collagenous fibers.
Fig. 1. A schematic illustration of acupuncture points, postulated by Niboyet (1979).
Fig. 2. Histological view of the acupuncture point of Hoku (LI-4) in the left foreleg of the mouse.
X70
Fig. 3. At the point of Hoku (LI-4) in the right foreleg of the mouse, one can also see a complex of nervous elements (N) and the vascular system (V). X70
Fig. 4 This light micrograph also shows a complex of nervous elements (N) and vascular system (V) in Taichong (LV-3) of the left hind leg of the mouse. X150
Fig. 5. There is also observed a complex of the nervous elements (N) and vascular system (V) in Taichong (LV-3) of the right hind leg of the mouse. X150
Fig. 6. A complex of the nervous elements (N) and vascular system (V) is also observed in an acupuncture point of the mouse auricle. X150
Fig. 7. A complex of the nervous elements and vascular system is not observed in the non-acupuncture point area of the mouse foreleg. X70