Development of the C1–C2 Puncture in Neuroradiology: A Historical Note
========================================================================

* E. Ralph Heinz

From the time of Ayer in 1920, the accepted method of accessing CSF from the craniocervical junction was the cisternal puncture, in which the spinal needle is directed sagittally in a midline plane from a point just beneath the occiput. For this approach, the patient was placed in the lateral decubitus position or seated upright in a chair with his or her head flexed. An assistant maintained the patient’s head in position. The needle was simply advanced until CSF was obtained, at which point the advance of the needle was stopped. The entire process was freehand. However, given the trajectory of the needle directed toward the vulnerable brainstem, the short distance between the dura and medulla, the possibility of head motion, and the absence of good monitoring technique, this method had marked limitations. It is not surprising that a number of complications occurred. Such complications included medullary injury, as evidenced by vomiting or cessation of breathing; venous or arterial perforation; and compromised vertebral blood flow. With these problems, there was obviously a need for a new route to the CSF in the high cervical region. Nonetheless, in the absence of a reliable alternative, cisternal punctures continued, at least until 1973 (1).

At that time, a number of compelling needs prompted the development of a procedure for high cervical puncture. From the 1940s through the early 1960s, pneumoencephalography and myelography were standard diagnostic tests. These required the needle tip to be solely in the subarachnoid space to allow the installation of air or contrast agent in that space rather than a mixed injection in which some of the injection went into the subdural space. While myelography was frequently successful with a limited mixed injection, an air study never was: The patient had to be discharged home and brought back for a repeat study after the subdural collection had disappeared. In a training program, this repetition happened fairly frequently. With new access to the subarachnoid space in the high cervical region, however, the study could go forward without delay.

Besides the need to avoid mixed injections, there were three other reasons for a high puncture. First was the need to access CSF at the skull base for bacterial culture in patients with meningitis and loculation or to obtain a sample of CSF near the brain to analyze for tumor cells. Second was a need to access CSF above a spinal block. Lumbar puncture below the block might well precipitate herniation of a spinal mass, leading to paraparesis because of the lowered CSF pressure in the lumbar region. Third was a need to allow the performance of painless gas myelography. While Pantopaque (Lafayette Pharmacal, IN) was a radioattenuated material widely used in the United States for myelography at the time, its use had many drawbacks, including arachnoiditis. In Scandinavia, the use of Pantopaque was avoided for decades, and gas myelography was preferred as a contrast study. Some used a lumbar spinal approach (2) in which the patient’s head was tilted sharply laterally toward the shoulder to keep air out of the head to minimize the adverse effects of headache and nausea. However, this technique did not always work and was not adopted in the United States.

If we were to obtain similar gas myelograms in the United States, how would we do so? We had some ideas. If we had our patients lie on a tomographic table with their head lower than their feet when the air was injected, we could titrate the exchange of gas for CSF, filling the entire lumbar and thoracic subarachnoid space, and still keep air out of the head. The gas could completely replace the CSF, and tomography could then be used to provide elegant sagittal radiographs of the spinal cord. To accomplish this, we punctured the subarachnoid space in the neck, allowing an exchange of gas for the CSF to the level of C1–2 that replaced all of the CSF throughout the lumbar, thoracic, and cervical regions *without* allowing air to reach the cranial subarachnoid space.

We noted that Rosomoff (3), a neurosurgeon, and colleagues showed that they could perform percutaneous cordotomy at C1–2 to relieve pain. If the neurosurgeons could safely perform an invasive procedure such as percutaneous cutting of the long tracts of the spinal cord, then we could perform the much less invasive procedure of accessing the CSF at that level, supplanting a cisternal midline puncture with a lateral cervical approach.

After studying the bony and soft tissue anatomy, including the course of the vertebral artery, we found that the subarachnoid space (although it was small) tended to open up posteriorly. One could puncture the subarachnoid space in the posterior part of the spinal canal from the direct lateral direction. With the needle point in position, the needle shaft would be held in alignment by the muscles in the lateral neck, so that with tubing attached and with the gentle aspiration of fluid, an injection of air could take place. The C1–2 lateral puncture could then be used for the exchange of gas for fluid in gas myelography, creating the so-called painless gas myelographic study. This technique was used in 1969–1970 at Yale University and was reported at the annual meeting of the Radiological Society of North America (RSNA) in November 1970 and in *Radiology* in 1972 (4).

In addition to gas myelography, this new high cervical puncture could be used to salvage a pneumoencephalogram or myelogram, to aspirate CSF for bacteriology, to detect malignant cells near the brain, or to perform a puncture to obtain CSF above a spinal block. The procedure has been used successfully throughout the neuroradiologic world.

Unknown to the author until years later, Dr David J. Kelly, Jr. and Dr. Eben Alexander, neurosurgeons at Bowman Gray Medical Center, Winston-Salem, NC, reported use of a lateral C1–2 approach to instill positive contrast material (Pantopaque), into the high cervical subarachnoid space in 1968 (5). However, the spread of knowledge about the C1–2 puncture to neuroradiologists was thought to occur primarily by means of the RSNA presentation, by the publication in *Radiology*, and by personal and phone conversations concerning the procedure with scores of neuroradiologists at the time.

## References

1.  Kendall B. **How to do a cisternal puncture.** Br J Hosp Med 1980;24:571
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=7272541&link_type=MED&atom=%2Fajnr%2F26%2F1%2F5.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=A1980KV34800010&link_type=ISI) 

2.  Roth M. **Gas myelography by the lumbar route.** Acta Radiol Diagn 1963;1:53–60
    
    

3.  Rosomoff HL, Carrol F, Brown J, Sheptal P. **Percutaneous radiofrequency cervical cordotomy: technique.** J Neurosurg 1965;23:639–644
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.3171/jns.1965.23.6.0639&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=5217032&link_type=MED&atom=%2Fajnr%2F26%2F1%2F5.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=A19657227700017&link_type=ISI) 

4.  Heinz ER, Goldman, RL. **The role of gas myelography in neuroradiologic diagnosis.** Radiology 1972;102:629–634
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.1148/102.3.629&link_type=DOI) 
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=5060173&link_type=MED&atom=%2Fajnr%2F26%2F1%2F5.atom) 
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=A1972L613600033&link_type=ISI) 

5.  Kelly DL Jr, Alexander E Jr. **Lateral cervical puncture for myelography.** J Neurosurg 1968;29:106–110
    
    [PubMed](http://www.ajnr.org/lookup/external-ref?access_num=5674086&link_type=MED&atom=%2Fajnr%2F26%2F1%2F5.atom) 

*   Copyright © American Society of Neuroradiology