Lasers in Surgery and Medicine 12:621-624 (1992)
Percutaneous Laser Discectomy With the Ho:YAG Laser Matthew R. Quigley, MD,Tommy Shih, MD,Amr Elrifai, MD, Joseph C. Maroon, MD, and Michael L. Lesiecki, PhD Allegheny General Hospital, Pittsburgh, Pennsylvania 15212, and Medical College of Pennsylvania, Philadelphia, Pennsylvania 19129 (M.R.Q., T.S.,A. E., J.C.M.), and Candela Laser Corporation, Wayland, Massachusetts 01778 (M.L.L.)
A prototype Ho:YAG (2.15 pm) laser operating at 2-J/pulse,3 Hz through a 600-pm fiber was employed to perform laser discectomies at the L3-4 disc through an 18G needle in five juvenile pigs. No temperature elevations were recorded in the posterior longitudinal ligament at the disc level and all animals recovered fully with no adverse sequelae, even immediately upon awakening from anesthesia. Pathologic examination demonstrated a wide swath of coagulation necrosis confined to the disc space. The Ho:YAG laser, owing to its close approximation to the intense 2.0 pm absorption band of water, appears to be a viable candidate for clinical trials of laser discectomy. o 1892 Wiley-Liss, Inc. Key words: lumbar disc, necrosis, pigs, pulsed infrared laser
changes in disc elastance were proportional to the mass-of nuclear material removed and that each The percutaneous treatment of lumbar disc treatment modality exhibited a plateau repredisease by the automated suction device (ASD) senting an optimal treatment parameter beyond has achieved widespread acceptance due to miniwhich no further disc removal was effected. We mal patient morbidity and brief hospital stays with reported success rates of 75%in selected pop- subsequently employed this plateau data t o guide ulations [ll. The ASD suffers from a number of our investigations performing in vivo percutaneous laser discectomy trial utilizing 2.15 pm Ho: shortcomings, however: 1)cumbersome three-step insertion procedure with large (2.1 mm) cannula; YAG laser in juvenile swine. 2) limited access to L5-S1 despite development of curved cannula; 3) disc removal limited to what is MATERIALS AND METHODS “suckable” [2]. Owing to these factors percutane- Absorption Spectrum ous discectomies through thin gauge needles have A human disc fragment excised from a 37been undertaken in limited clinical trials utilizyear-old patient undergoing a conventional miing the widely available Nd:YAG laser 131. Poor crodiscectomy procedure was suspended by glass water absorption at 1.06 pm wavelength, however, argues for implementation of lasers closer to slides lmm apart to obtain an infra-red absorpthe intense water absorption peak near 2.0 Fm tion spectrum utilizing a Fourier Transform Infrared Spectrometer (Digilab FTS-40, Cambridge, [41. MA). In a preliminary study we developed an in vitro model t o quantitate the mass of disc removal following various intradiscal treatments [51. In Accepted for publication August 15, 1992. brief, following needle cannulation of the disc, an Address reprint requests to Matthew R. Quigley, M.D., 420 infusion pump compresses a column of air within East North Avenue, East Wing, Suite 302, Pittsburgh, PA the needle, tubing and disc and the slope of the 15212. resultant line (pressure/volume) measured by a This investigation was supported by PHS research grant No. strain gauge is the elastance. We found that IR43-AR40278-01. INTRODUCTION
0 1992 Wiley-Liss, Inc.
622 Animal Studies
Quigley et al.
Five juvenile swine (30-40 kg) of mixed sexes were operated upon following thorough protocol review by the Institutional Care Committee for compliance with the Animal Welfare Act. After induction of a general anesthetic, pigs were intubated, maintained on 1%Halothane anesthesia and placed in a right decubitus position. In two animals, either the right or left side of the L3-4 disc was exposed by partial facetectomy and laminectomy to allow placement of a thermocouple (Omega Engineering, Stanford, CT) into the posterior longitudinal ligament beneath the nerve root. Measurements were therefore taken ventral t o the root both ipsilateral and contralatera1 to the side from which the laser was introduced. Three animals did not undergo this monitoring. Under fluoroscopic guidance a 5cm lateral paramedian approach was employed to cannulate the L3-4 disc with an 18G needle. Approximately 1cc of renografin-60 was hand injected into the disc to assure proper placement. A 600 pm fiber was advanced t o a predetermined depth into the disc space and 500 J of Ho:YAG1 energy (3 Hz, 2 J/pulse) applied in 100-5 increments (with 15-sec pauses between applications) while gently moving the fiber within the nucleus. These power and energy parameters were selected on the basis of disc elastance measurements previously reported [5]. The fiber and needle were removed, wounds closed and the animals allowed t o recover being observed for any untoward neurologic effects. At 1 week, animals were re-anesthetized and sacrificed. The entire L2-4 segment including spinal cordtroots was removed and representative samples of the L3-4 disc and adjacent cord sent for H&E analysis. RESULTS Absorption Spectrum
Discrete spectral features were observed in this analysis as indicated in Figure 1. The peaks near 1.5 pm and 2.0 pm correspond to characteristic water absorption bands [41, the primary constituent of disc [6]. The broad rise in absorbance at longer wavelengths (> 2.2 pm) also is attributable t o the intense water and collagen absorption near 3 pm, while the rise at shorter wave lengths (< 1.2 pm) is presumed secondary t o light scattering, especially by the collagen fibrils 0.1pm size [71. 'Prototype laser by Candela Laser Corp., Wayland, MA.
12-
0.8
-
0.4
-
0.2
.
Y"
28s rl
I ? O
12
14
16
18
20
22
24
26
Wavelength (microns)
Fig. 1. Absorption spectrum of human disc fragment obtained at surgery.
Animal Studies
The L3-4 disc space was easily cannulated in all swine, confirmed by discography (Fig. 2A). No annular temperature elevations were recorded in the two animals so monitored. Following treatment, the disc space by fluoroscopy assumed a cavitated appearance (Fig. 2B), presumably corresponding to the disc material removed. No animals evidenced any injury postoperatively, either neurologic or visceral. Post-mortem examination of the treated disc revealed a wide swath of necrotic material (Fig. 31, and there was no indication of injury extending beyond the disc annulus, even on microscopic sections. DISCUSSION
The disc absorption spectrum is relevant as it indicates the absorption target for the laser and in part dictates the choice of wavelength to be used. The histology of the disc has been well delineated in published studies-containing the avascular nucleus pulposis surrounded by the annulus fibrosis [6]. The nucleus is highly hydrophilic, containing 80438% water early in life, but diminishing to less than 70% in elderly individuals. The primary target of clinical laser disc ablation therefore is disc water. Percutaneous laser discectomy (PLD) is an intriguing but as of yet relatively untried method for treating lumbar disc disease. Limited clinical series have been reported employing the Nd:YAG laser with encouraging results [31. Inspection of the absorption spectrum of water, however, would suggest more efficient energyheat transfer utilizing a laser in the 2.0 pm range [41. The Ho:YAG
Ho:YAG Laser Discectomy
623
Fig. 2. a: Lateral discogram and artist’s rendition of L3-4 disc demonstrating filling of nucleus with dye. b: Same disc as in Figure 2a, post laser discectomy treatment. Notice cavitated appearance of disc.
was therefore employed in the in vivo model utilizing juvenile swine. Our selection of the 3-Hz, 2-J/pulse, 500 J total energy was based upon in vitro elastance measurements previously reported [ 5 ] . In that work we found in pig spine, following 500 J of Ho:YAG energy, maximal disc vaporization had occurred and that further energy increments resulted only in unwanted disc space heating. Pathological examination of the material from this experiment bore this out as
laser injury was confined solely to the disc space, where it was quite extensive. Temperature monitoring revealed no elevations during laser treatment. The thermocouples were placed immediately ventral t o the exiting L4 root both ipsi- and contralateral to the laser insertion site. It would be anticipated the roots at the treated level would be most susceptible t o laser energy. The absence of a temperature elevation adjacent to the annulus as measured by ther-
Quigley et al.
624
Fig. 3. Pathologic specimen of L3-4 disc a t sacrifice 1 week post laser discectomy. A band of necrotic disc material secondary to laser treatment is indicated by the arrows.
mocouples, however, also confirms the discrete nature of the Ho:YAG ablation. Future work in our laboratory will be directed at the determination of optimal energy/ power requirements in human disc as a preamble to clinical investigations.
2,
3.
ACKNOWLEDGMENTS
The authors wish to acknowledge David Compton for his assistance in the performance of disc absorption spectra, and Gina Thomas for her artistic contribution.
4. 5,
6.
REFERENCES 1. Onik G, Mooney V, Maroon JC, Wiltse L, Helms C, Schwiegel J, Watkinus R, Kahanovitz N, Day A, Morris J ,
7.
McCullough JA, Reicher M, Croissant P, Dunsker S , Davis AW, Brown C, Hochschuler S , Saul T, Ray C. Automated percutaneous discectomy: A prospective multi-institutional study. Neurosurgery 1990; 26228-233. Maroon JC, Onik G. Percutaneous automated discectomy: A new method for lumbar disc removal. J Neurosurg 1987; 66:143 -146. Choy DS, Case RB, Fielding W, Hughes J, Liebler W, Ascher P. Percutaneous laser nucleolysis of lumbar discs. N Engl J Med 1987; 317:771-772. Curcio JA, Petty CC. The near-infrared absorption specAm 1951; 41:302-304. trum of liquid water. J Opt SOC Quigley MR, Shih T, Elrifai A, Loesch DV, Maroon JC. Laser discectomy: Comparison of Ho:YAG and Nd:YAG systems. Surg Forum 1991; 42507-509. Humzah MD, Soames RW. Human intervertebral disc structure and function. Anat Rec 1988; 220:337-356. Nuss R, Fabian R, Sakar R, Puliafito C. Infrared laser bone ablation. Lasers Surg Med 1988; 8:381-389.
Percutaneous Laser Discectomy With the Ho:YAG Laser Matthew R. Quigley, MD,Tommy Shih, MD,Amr Elrifai, MD, Joseph C. Maroon, MD, and Michael L. Lesiecki, PhD Allegheny General Hospital, Pittsburgh, Pennsylvania 15212, and Medical College of Pennsylvania, Philadelphia, Pennsylvania 19129 (M.R.Q., T.S.,A. E., J.C.M.), and Candela Laser Corporation, Wayland, Massachusetts 01778 (M.L.L.)
A prototype Ho:YAG (2.15 pm) laser operating at 2-J/pulse,3 Hz through a 600-pm fiber was employed to perform laser discectomies at the L3-4 disc through an 18G needle in five juvenile pigs. No temperature elevations were recorded in the posterior longitudinal ligament at the disc level and all animals recovered fully with no adverse sequelae, even immediately upon awakening from anesthesia. Pathologic examination demonstrated a wide swath of coagulation necrosis confined to the disc space. The Ho:YAG laser, owing to its close approximation to the intense 2.0 pm absorption band of water, appears to be a viable candidate for clinical trials of laser discectomy. o 1892 Wiley-Liss, Inc. Key words: lumbar disc, necrosis, pigs, pulsed infrared laser
changes in disc elastance were proportional to the mass-of nuclear material removed and that each The percutaneous treatment of lumbar disc treatment modality exhibited a plateau repredisease by the automated suction device (ASD) senting an optimal treatment parameter beyond has achieved widespread acceptance due to miniwhich no further disc removal was effected. We mal patient morbidity and brief hospital stays with reported success rates of 75%in selected pop- subsequently employed this plateau data t o guide ulations [ll. The ASD suffers from a number of our investigations performing in vivo percutaneous laser discectomy trial utilizing 2.15 pm Ho: shortcomings, however: 1)cumbersome three-step insertion procedure with large (2.1 mm) cannula; YAG laser in juvenile swine. 2) limited access to L5-S1 despite development of curved cannula; 3) disc removal limited to what is MATERIALS AND METHODS “suckable” [2]. Owing to these factors percutane- Absorption Spectrum ous discectomies through thin gauge needles have A human disc fragment excised from a 37been undertaken in limited clinical trials utilizyear-old patient undergoing a conventional miing the widely available Nd:YAG laser 131. Poor crodiscectomy procedure was suspended by glass water absorption at 1.06 pm wavelength, however, argues for implementation of lasers closer to slides lmm apart to obtain an infra-red absorpthe intense water absorption peak near 2.0 Fm tion spectrum utilizing a Fourier Transform Infrared Spectrometer (Digilab FTS-40, Cambridge, [41. MA). In a preliminary study we developed an in vitro model t o quantitate the mass of disc removal following various intradiscal treatments [51. In Accepted for publication August 15, 1992. brief, following needle cannulation of the disc, an Address reprint requests to Matthew R. Quigley, M.D., 420 infusion pump compresses a column of air within East North Avenue, East Wing, Suite 302, Pittsburgh, PA the needle, tubing and disc and the slope of the 15212. resultant line (pressure/volume) measured by a This investigation was supported by PHS research grant No. strain gauge is the elastance. We found that IR43-AR40278-01. INTRODUCTION
0 1992 Wiley-Liss, Inc.
622 Animal Studies
Quigley et al.
Five juvenile swine (30-40 kg) of mixed sexes were operated upon following thorough protocol review by the Institutional Care Committee for compliance with the Animal Welfare Act. After induction of a general anesthetic, pigs were intubated, maintained on 1%Halothane anesthesia and placed in a right decubitus position. In two animals, either the right or left side of the L3-4 disc was exposed by partial facetectomy and laminectomy to allow placement of a thermocouple (Omega Engineering, Stanford, CT) into the posterior longitudinal ligament beneath the nerve root. Measurements were therefore taken ventral t o the root both ipsilateral and contralatera1 to the side from which the laser was introduced. Three animals did not undergo this monitoring. Under fluoroscopic guidance a 5cm lateral paramedian approach was employed to cannulate the L3-4 disc with an 18G needle. Approximately 1cc of renografin-60 was hand injected into the disc to assure proper placement. A 600 pm fiber was advanced t o a predetermined depth into the disc space and 500 J of Ho:YAG1 energy (3 Hz, 2 J/pulse) applied in 100-5 increments (with 15-sec pauses between applications) while gently moving the fiber within the nucleus. These power and energy parameters were selected on the basis of disc elastance measurements previously reported [5]. The fiber and needle were removed, wounds closed and the animals allowed t o recover being observed for any untoward neurologic effects. At 1 week, animals were re-anesthetized and sacrificed. The entire L2-4 segment including spinal cordtroots was removed and representative samples of the L3-4 disc and adjacent cord sent for H&E analysis. RESULTS Absorption Spectrum
Discrete spectral features were observed in this analysis as indicated in Figure 1. The peaks near 1.5 pm and 2.0 pm correspond to characteristic water absorption bands [41, the primary constituent of disc [6]. The broad rise in absorbance at longer wavelengths (> 2.2 pm) also is attributable t o the intense water and collagen absorption near 3 pm, while the rise at shorter wave lengths (< 1.2 pm) is presumed secondary t o light scattering, especially by the collagen fibrils 0.1pm size [71. 'Prototype laser by Candela Laser Corp., Wayland, MA.
12-
0.8
-
0.4
-
0.2
.
Y"
28s rl
I ? O
12
14
16
18
20
22
24
26
Wavelength (microns)
Fig. 1. Absorption spectrum of human disc fragment obtained at surgery.
Animal Studies
The L3-4 disc space was easily cannulated in all swine, confirmed by discography (Fig. 2A). No annular temperature elevations were recorded in the two animals so monitored. Following treatment, the disc space by fluoroscopy assumed a cavitated appearance (Fig. 2B), presumably corresponding to the disc material removed. No animals evidenced any injury postoperatively, either neurologic or visceral. Post-mortem examination of the treated disc revealed a wide swath of necrotic material (Fig. 31, and there was no indication of injury extending beyond the disc annulus, even on microscopic sections. DISCUSSION
The disc absorption spectrum is relevant as it indicates the absorption target for the laser and in part dictates the choice of wavelength to be used. The histology of the disc has been well delineated in published studies-containing the avascular nucleus pulposis surrounded by the annulus fibrosis [6]. The nucleus is highly hydrophilic, containing 80438% water early in life, but diminishing to less than 70% in elderly individuals. The primary target of clinical laser disc ablation therefore is disc water. Percutaneous laser discectomy (PLD) is an intriguing but as of yet relatively untried method for treating lumbar disc disease. Limited clinical series have been reported employing the Nd:YAG laser with encouraging results [31. Inspection of the absorption spectrum of water, however, would suggest more efficient energyheat transfer utilizing a laser in the 2.0 pm range [41. The Ho:YAG
Ho:YAG Laser Discectomy
623
Fig. 2. a: Lateral discogram and artist’s rendition of L3-4 disc demonstrating filling of nucleus with dye. b: Same disc as in Figure 2a, post laser discectomy treatment. Notice cavitated appearance of disc.
was therefore employed in the in vivo model utilizing juvenile swine. Our selection of the 3-Hz, 2-J/pulse, 500 J total energy was based upon in vitro elastance measurements previously reported [ 5 ] . In that work we found in pig spine, following 500 J of Ho:YAG energy, maximal disc vaporization had occurred and that further energy increments resulted only in unwanted disc space heating. Pathological examination of the material from this experiment bore this out as
laser injury was confined solely to the disc space, where it was quite extensive. Temperature monitoring revealed no elevations during laser treatment. The thermocouples were placed immediately ventral t o the exiting L4 root both ipsi- and contralateral to the laser insertion site. It would be anticipated the roots at the treated level would be most susceptible t o laser energy. The absence of a temperature elevation adjacent to the annulus as measured by ther-
Quigley et al.
624
Fig. 3. Pathologic specimen of L3-4 disc a t sacrifice 1 week post laser discectomy. A band of necrotic disc material secondary to laser treatment is indicated by the arrows.
mocouples, however, also confirms the discrete nature of the Ho:YAG ablation. Future work in our laboratory will be directed at the determination of optimal energy/ power requirements in human disc as a preamble to clinical investigations.
2,
3.
ACKNOWLEDGMENTS
The authors wish to acknowledge David Compton for his assistance in the performance of disc absorption spectra, and Gina Thomas for her artistic contribution.
4. 5,
6.
REFERENCES 1. Onik G, Mooney V, Maroon JC, Wiltse L, Helms C, Schwiegel J, Watkinus R, Kahanovitz N, Day A, Morris J ,
7.
McCullough JA, Reicher M, Croissant P, Dunsker S , Davis AW, Brown C, Hochschuler S , Saul T, Ray C. Automated percutaneous discectomy: A prospective multi-institutional study. Neurosurgery 1990; 26228-233. Maroon JC, Onik G. Percutaneous automated discectomy: A new method for lumbar disc removal. J Neurosurg 1987; 66:143 -146. Choy DS, Case RB, Fielding W, Hughes J, Liebler W, Ascher P. Percutaneous laser nucleolysis of lumbar discs. N Engl J Med 1987; 317:771-772. Curcio JA, Petty CC. The near-infrared absorption specAm 1951; 41:302-304. trum of liquid water. J Opt SOC Quigley MR, Shih T, Elrifai A, Loesch DV, Maroon JC. Laser discectomy: Comparison of Ho:YAG and Nd:YAG systems. Surg Forum 1991; 42507-509. Humzah MD, Soames RW. Human intervertebral disc structure and function. Anat Rec 1988; 220:337-356. Nuss R, Fabian R, Sakar R, Puliafito C. Infrared laser bone ablation. Lasers Surg Med 1988; 8:381-389.
