Complications after Artificial Disc Replacement

Analysis of cases between 2014 and 2022 - Stenum Hospital

There is no medical therapy that works without complications or side effects. Therefore, the objective consideration of occurring complications is imperative.

In the case of complications, the question always arises as to what caused the problems. Is the error to be found in the indication? Is the choice of suitable drugs or implants faulty? Was the treatment provided without error, in accordance with current scientific standards?

All these questions require thorough consideration in every medical field, including the field of intervertebral disc prosthetics - a treatment method that has been established for 40 years, but is also the subject of enormous controversy.

In Germany, all medical procedures performed in hospitals licensed as planned hospitals are centrally recorded statistically, and both the diagnoses and the procedures of the treatment are processed via coding systems (ICD and OPS).

From these data, which are collected and published by the Federal Statistical Office, it can be determined that approximately 6000 operations with artificial disc replacement (disc endoprosthesis) are performed in Germany per year. Any revisions are also recorded using special OPS codes and this rate is approximately 5% per year.

For comparison, the revision rate for pure decompressions is on average about twice as high.

In the scientific literature, the complication and revision rates for disc arthroplasty and fusion surgery, respectively, are as follows:

 

	Total disc replacement	fusion
peri- and postoperative complications	2-8%	4-18%
Revision surgery	2-4%	3-12%
clinical success	>85%	<90%
adjacent level disease	<2%	>15%

 

Actually, these figures already show a clear superiority of disc prosthetics compared to fusion surgery. Nevertheless, disc prosthetics is clearly criticized.

There are several reasons for this. First, disc prosthetics of the lumbar spine is linked to the availability of appropriate experience and knowledge of the surgical approach. Unfortunately, only a few surgeons worldwide have profound experience in the field of surgical access. In addition, experience in making indications also plays a significant role - if the procedure is not routinely performed, experience in making indications is also lacking. In addition, the assessment of postoperative results is enormously dependent on experience. Patients are usually chronic back pain patients with histories of several years. As a rule, all of these patients will have postoperative complaints, regardless of whether they underwent fusion or disc prosthesis. However, the assessment of fused patients seems to be easier - after all, fusion is the last ultima ratio and what has been fixed can then no longer cause any complaints. In contrast, the disc prosthesis offers the chance of preserving motion, and if the patients continue to have complaints postoperatively, the

 

prostheses or the facets are quickly the "culprits". This regularly ignores the fact that complications also occur after fusion surgery.

The following statistics are based on cases at the Stenum Orthopaedic Clinic from 2014 to 2022 and thus represent approximately 20% of the cases provided by my team at the Potsdam and Stenum sites, among others, over the past 20 years.

Between 2014 and 2022, we provided disc prosthetics to 2614 patients in the cervical and lumbar spine, and we fused 2417 patients during the same period. Pre-operations and revisions were performed on 1643 patients. This means that minus our own revisions, every third patient who reached the clinic had already been pre-operated on.

In this paper, I would like to focus on the patients who underwent disc prosthesis and try to name the failure analysis for failure.

In the context of the 2614 patients treated with disc prostheses, we had 169 patients with revision operations after disc prosthesis in the same period. This means that the revision rate is 2.75%. In relation to the number of implanted prostheses (4265), the revision rate is 1.69%.

But what were the reasons for a revision? Here, of course, several considerations must be made:

1. patients who were fitted ex domo or in domo prior to revision
2. analysis of the reasons for revision
3. analysis of the implant types

Only this overall view can help us to recognize certain patterns and this in turn is helpful to avoid errors in the future.

The following errors could be systematically identified:

• Hematoma early/postoperatively
• subsidence
• migration
• Root irritation early
• Early complication SPL
• Paresis early
• early infection
• Defect of the prosthesis
• low grade infection
• Spondylolisthesis aquisata
• osteolysis
• Stick-slip friction
• hypermobility
• instability

 

• Rotational Slippage
• loosening
• persistent stenosis
• Late acute infection

Here again, a distinction must be made between early and late complications. The early complications were postoperative hematoma, nerve root irritation, and paresis, followed by subsidence, migration, and the early complication spondylolisthesis aquisata as immediate postoperative instability. Only one early infection was noted.

Late complications were prosthesis defects and low grade infections followed by spondylolisthesis aquisata, osteolysis of unclear cause, stick-slip friction, hypermobility, instability, rotational slippage, loosening, persistent stenosis and, in distinction from low grade infections, acute late infection.

 

reason of revision	in domo	Implant type	ex domo	Implant type
defect			67	Cadisk-C und -L
low grade infection	25	M6-C 19 M6-L 5 Triadyme 1	6	M6-C 5 ESP-C 1
Spndylolisthisis aquisata	16	M6-L 14 M6-C 1 Active-L 1
Osteolysis			10	ESP-L
Stick-Slip Friction	10	Triadyme
subsidence	5	M6-C 2 M6-L 3	3	Active-C 1 M6-L 2
Hematoma early	5	M6-C 4 Triadyme 1
migration	2	M6-L	2	M6-L 1 ESP-L 1
Hypermobility	1	M6-C 1	3	M6-L 1 Mobi-C 1 Rotaio 1
Instability	1	M6-L 1	3	M6-L 1 Mobi-C 2
Nerve root irritation early	3	Baguera-L 1 M6-C 1 M6-L 1
Rotational slippage			2	M6-L 2
loosening			2	DCI 1 ESP-C 1
persistent stenosis	1	M6-C 1	1	M6-C 1
Late infection acut	1	M6-L 1	1	M6-C
Early infection	1	M6-L
Paresis early	1	M6-C 1
Total	72		100

 

If the reasons for revision are further differentiated, it becomes apparent that, as a rule, early complications are not implant-associated.

In the case of late complications, there are clear indications of implant-associated and non-implant- associated complications.

Thus, all cases of prosthesis defects are exclusive to Cadisk-C and -L.

Low-grade infections occur with different prosthesis types, but in our group there is a high incidence with the M6-C, although it must be emphasized that we also have very high numbers of implants here. In the context of the scientific literature, however, it should be noted that the affinity of bacteria for plastics is significantly higher than for metals. If bacteria enter the bloodstream (hematogenous infection pathway), the probability of bacterial colonization of a plastic surface is greater than with a metallic surface. Nevertheless, the numbers are in the absolute per mille range.

We have seen osteolyses of unclear cause exclusively in ESP-L prostheses, usually occurring 4-5 years after implantation and again leading to increasing complaints. The reason for their occurrence is unclear, but it must be assumed that they only occur with one type of prosthesis, so that there is a clear implant association here.

The same applies to the stick slip friction. We have observed this effect exclusively with the Triadyme prosthesis. The spherical design of the surface can lead to overloading of the articulating surfaces with perceptible crepitation during rotational movements of the head.

Subsidence, in turn, affects different types of prosthesis and is not causally implant-associated; rather, due to the force distribution occurring immediately after surgery, occasional weakening of the cover plate of the caudal vertebra occurs with sinking of the prosthesis.

In the migrations that have occurred, the L5/S1 segments are usually affected. In this case, it is not the implant that is the cause, but rather the implantation technique.

The complex of hypermobility, instability and rotational slippage can essentially be summarized as increasing instability of the affected segment, which the prosthesis must follow due to its design.

Intervertebral disc prostheses are not designed to compensate for segmental instabilities that occur, which is why primary implantation is contraindicated in cases of known instability. However, it cannot be ruled out that instabilities may develop over time even in supplied segments, which then overtax the lying implant.

The situation is similar with spondylolisthesis aquisata; this pathology is already known after fusion surgery. The present cases almost exclusively involve the L5/S1 segment and almost without exception the M6-L. Since the prosthesis, in contrast to classical ball and socket prostheses, has all the movement qualities of a normal healthy intervertebral disc, the possibility of translation is given and thus the prosthesis will passively follow a sliding of the respective cranial vertebra and thus be damaged.

Persistent stenoses are the result of insufficient primary decompression or arise on the basis of secondary hypersostotic bone formation postoperatively. In either case, they are not implant associated.

Acute late infections must be distinguished from low grade infections. These arise as acute, clinically imposing infections with accompanying abscess formation and are also known from other areas of orthopedic surgery. As with low grade infections, the pathway of origin is also hematogenous. Again, there is no implant-associated origin.

 

In summary, the present analysis shows that 87 of 172 cases are implant associated. Except for 10 cases, all other cases were pretreated in other hospitals. No implant-associated complication was found in any case involving the M6 prosthesis.

During the same period, a total of 4265 prostheses were implanted in the cervical and lumbar spine in 2614 patients. Related to these numbers, 72 revision surgeries were necessary for various reasons. Thus, the number of necessary revisions is 2.75% in relation to the number of patients and 1.69% in relation to the number of implants. Overall, the revision rate is thus about half of the German average. Only 0.38% of the revisions were implant associated. All other revisions showed no implant associated causes.

The present analysis considers cases that were surgically treated between 2014 and 2022. However, in total, the number of patients we cared for is much higher and is likely to approximate 10,000 cases. Thus, further workup will continue, but is time-consuming given the large number.