Several of our patients have insisted upon going home on the tenth day, while others who were ready to go home on the twelfth day, for one cause or another, found it inconvenient to leave at that time, or else they lived at a distance and they were advised not to attempt to travel so soon. In the uninfected cases, the blood and liquor amnii which finds its way into the peritoneal cavity does no harm and no great effort is made to remove it.

We find the following advantages in the use of the small median incision entirely above the umbilicus: There is no danger of adhesions between the uterine and the abdominal wounds, and the uterus is therefore allowed to involute normally and take up its position in the pelvis without restricted mobility.

In the midline the abdominal wall is very thin; no important structures are divided and the tissues are quite elastic so that a small opening is all that is necessary for the delivery of the child. The small abdominal opening offers much less chance for the escape of intestine and omentum and less opportunity and necessity to handle the abdominal contents. Located above the umbilicus there is much less probability of the subsequent occurrence of hernia through the cicatrix, for it is above the most dependent part of the abdomen which is subjected to the greatest strain when the patient is in the upright position. More support is also given at this point by the recti muscles as they tend to come together toward their upper attachments.

While we have not had an opportunity to examine all of the patients upon whom we have performed this method of Cesarean section, yet there is a considerable number who return for subsequent delivery in this way, or for other causes, and thus far we have not seen a hernia in any of our patients, except in one woman in whom the entire thickness of the abdominal wound broke down, and was closed by granulation. This patient developed a hernia several months later through the site of her Cesarean wound in the abdomen.

THE CULTIVATION OF HUMAN TISSUE IN VITRO.1

By JOSEPH R. LOSEE, M.D.,
Pathologist to the Hospital

and

ALBERT H. EBELING.

(From the Laboratories of the New York Lying-In Hospital and of The Rockefeller Institute for Medical Research.)

THE present investigations were undertaken to ascertain whether human connective tissue taken from a fresh cadaver could be kept in a condition of permanent life outside of the organism. We have applied to human tissues the method by which Carrel was able to keep animal connective tissue alive in vitro for more than two years.2

The first attempt to cultivate human tissues in vitro was made in 1911 by Carrel and Burrows.3 Small fragments of human malignant tumors were placed in human plasma and incubated. In a few cases the fragments were surrounded after a few days by many cells; but generally liquefaction of the medium occurred and no growth was observed. In other experiments undertaken later by Carrel on the growth of normal tissues the same phenomenon was observed. Fragments of thyroid gland and fragments of connective tissue from adult individuals or from fetuses were inoculated into culture media which contained or did not contain tissue extract. The liquefaction was observed in this case in the same manner as in the case of the malignant tumors, and with the exception of a few cultures no real growth was observed. In 1913 Maccabruni cultivated in plasma the tissues of the uterus and ovary of human fetuses and observed around the tissue after 8 or 9 days the presence of many cells and also of karyokinetic figures. The result of these experiments showed that human tissue almost always produced liquefaction of the plasma, and that this probably prevented its growth. We, therefore, attempted to develop a technique which would permit us to keep human tissue in a plasmatic medium without the occurrence of liquefaction.

At first we attempted to obtain a medium which would not liquefy under the influence of the tissue. The first medium tried was composed of human plasma and extract of human tissue taken from fresh cadavers. 1. These investigations were carried on in March, 1913, in the research laboratory of the New York Lying-In Hospital. They were rendered possible by the cooperation of Dr. Markoe, of the attending surgeons of the Lying-In Hospital, and of Dr. Carrel. The Directors of the Hospital provided a specially equipped laboratory for the work, under the supervision of Dr. Markoe and Dr. Carrel, thus enabling us to carry out the study of the tissues of human cadavers under excellent conditions. Appears also in Jour. Exper. Med., Vol. xix, No. 6.

2. Carrel, A., Jour. Exper. Med., 1913, xviii, 287.

3. Carrel, A., and Burrows, M. T., idem, 1911, xiii, 387.

*Maccabruni, F., Ann. d. ostet. e ginecol., 1914, xxxvi, 57.

The addition

This, however, proved unsuitable, as in twenty-four hours liquefaction occurred about the fragments of tissue. Very little growth was observed, but when it took place the cells were seen to be scattered and the new growth to be very thin. A thin film containing many cells adhered to the cover-glass. When attempts were made to extirpate the film and to transfer it to fresh medium it generally became folded up and after one or two passages no further growth occurred. Many modifications of the medium were tried, in order to overcome these difficulties. Human plasma was diluted with agar, serum, and egg albumen. of agar and egg albumen to the plasma yielded a firmer medium. No liquefaction took place, but on the other hand no real growth occurred. When the plasma was diluted with serum complete liquefaction rapidly followed. The addition of a small amount of acid or alkali to the plasma did not prevent this liquefaction. Finally after many attempts had been made it was found that by diluting the plasma with equal parts of Ringer's solution a medium could be obtained which would not liquefy in less than 24 hours, and often not in 48 and 72 hours. Usually 18 hours after the medium had been inoculated with human tissue, growth appeared and increased progressively. After a period of from 24 to 96 hours the fragments of tissue were transferred to a fresh medium in which the growth continued. The medium was again modified by the addition of a small quantity of diluted extract of human tissue, after which the growth became very active. Finally at the end of 1913 and during the first months of 1914 it became possible to obtain large growths of human connective tissue. The tissue could be transferred from medium to medium; the new cells could be isolated from a primitive fragment and cultivated in a new medium; the mass of newly formed tissue rapidly increased in the same way that was observed by Carrel to occur in the case of the strain of connective tissue cells taken from the chick embryo, which are still proliferating at the Rockefeller Institute. Nevertheless, the technique which we have developed is still far from perfect. Liquefaction of the medium occurs as soon as the tissue extract becomes too concentrated. Besides, it appears that the plasma taken from many different individuals varies widely, the plasma of certain persons liquefying much more rapidly than that of others. The value of the extracts varies also, the result being that the medium is inconstant; and for this reason the rate of growth of fragments of tissue shows marked fluctuations in the course of a few weeks. The medium will probably be still further modified; nevertheless we shall describe here the technique that has rendered possible the cultivation of human tissue in vitro.

TECHNIQUE.

Plasma. Blood is aspirated from the median basilic vein of adults under aseptic conditions with a dry, chilled, 10 c.c. glass syringe (gauge of

needle, 19). The blood is immediately transferred to cold paraffined glass tubes and centrifuged in ice at high speed (3,000 revolutions) for 5 minutes. The plasma is at once pipetted off with a paraffined pipette into another cold paraffined tube, which was kept on ice ready for use. Several paraffined pipettes are prepared with bulbs to be kept in readiness, because it has been found that although the plasma remains liquid for some time in the paraffined tube it coagulates in the pipette used for making the cultures in from two to ten minutes. It is usually necessary to use a fresh pipette after changing two or three cultures.

Extract. The extract is prepared from tissues taken from adult or fetal cadavers, and cut into small pieces. Equal parts of Ringer's solution are added, the whole is placed in cold storage for 48 hours, after which it is centrifuged and the supernatant fluid pipetted off. This fluid may then be kept for some time in cold storage.

Making of Cultures.-Pieces of the various fetal tissue taken from fresh fetal cadavers are placed in Ringer's solution or in serum and cut to the proper size with a cataract knife. The plasma is first spread on the cover-glass, and equal parts of Ringer's solution are added and thoroughly mixed with it. The object of this procedure is to obtain a clot which can adhere firmly to the cover-glass, because it was found that when the plasma was diluted with the Ringer's solution first coagulation took place too rapidly, thereby preventing a thorough embedding of the tissue. On the other hand, when the Ringer's solution is spread first and the plasma added afterwards the clot does not adhere firmly, often contracting to such an extent as to become separated from the cover-glass, thereby destroying the culture. Then the tissue is put into the medium and the extract immediately added. The medium usually coagulates about the piece within a few seconds. A reasonable amount of speed and strict aseptic precautions are observed throughout. The preparations are next placed over a concave slide hermetically sealed with paraffin and incubated at 38°C., and the tissues allowed to grow. When a fresh amount of growth has taken place after a period varying from 24 to 96 hours, the tissues are removed from their medium, washed, and transferred to a fresh medium. The time of the transfer depends upon the condition of the medium and the activity of the growth. Observations are taken at intervals of 12, 24, 48, and even 96 hours. Observations are likewise made of cultures fixed and stained with hematoxylin and with Giemsa stain.

RESULTS.

With the technique described above, it became possible to keep a strain of connective tissue cells, derived from a piece of skin which was obtained from a fresh four months old fetal cadaver, in a condition of active life in vitro for more than two months.

When the fragments of fetal heart tissue, obtained from fresh cadavers,

were introduced into a medium composed of equal parts of human plasma and Ringer's solution, and incubated at 38°C. for 16 hours, macroscopic examination of the culture showed the fragments as sharply outlined masses in a slightly opalescent medium with no evidences of cell proliferation. In from 18 to 24 hours cell proliferation manifested itself, in some cases by the appearance of many cells spreading out or budding from the original piece; in other cases by the presence of only a few scattered cells.

The rate and extent of growth varied and probably depended on the vitality of the tissue, the constituents of the medium, and the changes that developed in it. Where there was a tendency for the proliferating cells to grow in one or two planes the growth was greater and more rapid than where the cells appeared to grow in several planes. In the latter case the growth was also more dense.

The phenomenon of growth was accompanied by a progressive and slow liquefaction of the medium around the primitive fragment and in the zone of cell proliferation. Sometimes liquefaction around the entire fragment took place within 24 hours after the cultures were made, although little, if any, growth was apparent. At times the medium retracted from the piece in certain areas and the vacuoles in the medium thus produced contained fluid in which detached cells and debris were observed. The liquefaction did not indicate necrosis of the piece of tissue in all instances. Pieces around which the medium liquefied completely in 24 hours have been transplanted into fresh medium, in which cell proliferation took place in from 16 to 24 hours. There was no evidence of rapid liquefaction.

When liquefaction was apparently due to cell proliferation, it started from the periphery of the fragment and was progressive. On the other hand, when apparently due to substances in the medium it was general and affected the entire mass of the medium. Liquefaction of the entire medium took place when extract from fetal tissues was used in primitive cultures. Primitive fragments grew well without the addition of extract to the medium for one or two passages, but usually after the second passage it was necessary to add extract in order to stimulate growth. The time of the passages into fresh medium was governed by the condition of the culture. Where the medium showed liquefaction around the primitive piece in 24 hours, a change was made. If the rate of growth was rapid, the culture was changed after 48 hours. Actively growing cultures which were not changed until after 72 hours grew less well. In cases where the time of passage was deferred for 96 hours, subsequent passages showed a decided retardation in the rate and extent of growth of the cells and the cellular characters were altered. Ameboid cells occurred and rapid liquefaction of the entire medium often took place; and although passages into fresh medium were made every 24 or 48 hours the cultures did not recover.

In the history of one culture, which is given in Table I, it will be