LP shunt vs. VP shunt

Since the 1950s, the implantation of a shunt system has been the standard therapy for the treatment of hydrocephalus. The shunt consists of a tube-valve system that drains excess cerebrospinal fluid (CSF) into another body cavity where it can be reabsorbed. At the same time, there have been and continue to be attempts to develop alternative treatments, for example, by means of drugs or by special surgical interventions, such as endoscopic third ventriculostomy (ETV) or partial or complete removal of the choroid plexus, the structures in the brain where cerebrospinal fluid is formed (1-5). However, these have ultimately remained without major relevance to date and have not been able to replace shunt therapy.

Implantation of a shunt is generally neither dangerous nor difficult compared to other neurosurgical procedures. A typical drainage system (see below) consists of at least silicone catheters and a valve that regulates intracranial pressure. The catheters, valve, and other optional components are implanted subcutaneously, meaning under the skin. The catheters are passed under the skin with a tunneling instrument, and the valve is fixed in a skin pocket using a small incision. Only the tips (openings) of the inflow and outflow catheters invasively penetrate the respective body cavities, depending on the type of shunt.

Types of shunt derivation: VP, LP, VA

Theoretically, a large number of body cavities can be considered for derivation. However, only three of these drainage options have any practical significance. All other types of drainage are used only in rare special cases.

1. ventriculoperitoneal (VP): from the cerebral ventricles (head) into the peritoneum (abdominal cavity)

2. lumboperitoneal (LP): out of the spinal canal into the peritoneum.

3. ventriculoatrial (VA): from the brain ventricles (head) into the right atrium of the heart.

These three types of derivation have been used since the 1960s and are well established operations. As shown in Figure 1, the valves and other optional components are placed retroauricularly (i.e., behind the ear), in the thoracic or in the lumbar region, among others.

At first glance, the LP shunt has some advantages. Although rarely practiced, this derivation option can be performed under local anesthesia if it offers advantages for the patient. Another particularly appealing aspect is that, in contrast to the VP shunt, the skull bone does not have to be pierced and therefore there is no risk of brain injury or even brain hemorrhage during shunt placement (17). Overall, LP shunt placement can be considered less invasive compared with VP shunt placement. However, for quite some time, LP shunting was generally "suspected" to be associated with unacceptably high complication rates compared with VP shunts (18-20). Larger, also randomized studies and meta-reviews presented in the last decade have been able to dispel this suspicion though (21-23). A good recent example is a meta-review by Giordan et al. who, on the basis of extensive data from more than 2,000 patients, demonstrated (at least) the equivalence of LP and VP shunts in terms of safety and performance, i.e., complication rates and therapeutic success ("outcome") (cf. Figure 2) (7).  

Such positive findings have led to an increase in the rate of LP shunts, especially in the Asian region such as Japan, but also worldwide. Already in 2011, more iNPH patients in Japan were treated with LP shunts than with VP shunts (21,24). Figure 3 reflects the increasing use of LP shunts worldwide through a literature search for the keyword "lumboperitoneal shunt" in the interdisciplinary "Web of science" (25).

The possible complications and their causes overlap only partially between VP and LP shunts. In addition to commonalities, such as mechanical valve problems, each shunt type is also associated with specific complications. Jusue-Torres et al. analyzed the different LP complication causes from 602 LP shunt studies between 1971 and 2013 based on 2,871 treated patients (25). The result (reduced to the main factors) is shown in the following Figure 4. The percentages refer to the cumulative total number of all complications.

"Although rare, over drainage causing slit ventricle syndrome, intracranial hypotension syndrome, chronic subdural effusion, or subdural hemorrhage may also be the disadvantages of VPS.

On the other hand, the lumbar exit and peritoneal entry for LPS are generally at the same level when the patients are upright to minimize the effect of gravity, and the siphoning effect is negligible."

In the same vein, Miyake writes (26):

"All valves except valves with a membrane-type antisiphon mechanism can be implanted in any location in LP shunts.

Use of valves with membrane type antisiphon mechanisms is unnecessary because the siphon effect is negligible in LP shunts."

The statement that the siphon effect is negligible if the inflow site (lumbar) and the outflow site (peritoneal) are at the same level in the standing position is formally correct, but seems misleading. The impression is created here that the hydrostatic pressure conditions in the standing position are irrelevant for the LP shunt and that no countermeasure (as in the VP shunt) is required. In the case of the LP shunt, although there is indeed no longer a "hanging column of fluid" BELOW the valve, the entire weight of the fluid column is present ABOVE the valve. This statement may be confusing, since there is no fluid-filled catheter at the top (as in the VP shunt). However, the same hydrostatic pressure exists in the (fluid-filled) spinal canal, which increases downward in proportion to the distance from the ventricles ("p =r *g*h"). However, for the valve, which is always a differential pressure valve, only the difference between the pressures at the inlet and outlet plays the decisive role. Thus, it does not matter whether the valve is opened while standing by a strong additional suction at the outlet ("siphoning") or by an increased pressure at the inlet. Therefore, the following formulation by Mirone et al. is more appropriate (29):

"Lumboperitoneal shunts may produce similar posture-related problems. The difference is that the hydrostatic pressure does not produce pulling but rather pushing force in the vertical body position."

The fact remains that with both the VP and LP shunt, an additional, strong differential pressure must be taken into account, which only occurs in the upright position. This unavoidably opens the valve and can thus lead to overdrainage. The hydrostatic pressure conditions are shown in Figure 5.

Various authors have shown that the implantation of an anti-siphon device (ASD) is therefore also necessary and effective in the LP shunt. Various gravitational valves and the so-called SiphonGuard have been used for this purpose (23,28,30). In a standing position, the SiphonGuard does not completely prevent (over)drainage, but it reduces the flow and thus slows down its development. In addition, the use of flow restrictors in the form of narrow lumen catheters (inner diameter 0.8 mm) to prevent overdrainage has also been described. These generally reduce flow in the LP shunt, i.e., regardless of body position. However, a recent comparative study by Nakajima with 115 included iNPH patients suggests a superiority of gravitational valves (30). The membrane-based ASDs mentioned by Miyake (Delta® Chamber & ASD) represent a special case, as they respond differently to pressure at the inlet or suction at the outlet. Their function depends on the appropriate inlet pressure/outlet suction ratio, and thus essentially on the appropriate implantation height in the body. Therefore, such membrane valves can indeed to be considered problematic for LP shunts (26).

Figure 6 shows a schematic comparison of the dynamic relationships within the VP and LP shunts. In the VP shunt, CSF is drained directly from the ventricles into the abdomen. If slit ventricles form due to overdrainage, the ventricular catheter, which is then trapped by tissue, is particularly vulnerable to occlusion. Due to the different type of drainage, this is not possible with the LP shunt. Here, the excess CSF first flows through the IV. ventricle into the subarachnoid and then into the spinal space before being drained from there into the abdominal cavity. Obstruction of the lumbar catheter by tissue entrapment is less likely here. However, the flow movement from the upper three ventricles through the IV. ventricle into the spinal canal carries the risk of entrainment/displacement of certain brain parts, which can lead to equally dangerous herniation (see flow direction LP shunt Fig. 6). This relationship becomes very clear in non-communicating hydrocephalus, especially due to an occluded aqueduct. The overpressure prevailing in the upper three ventricles would move the whole brain towards the spinal canal as soon as the spinal canal would be opened. For this reason, non-communicating hydrocephalus is also an important contraindication for LP shunting (13).

As described, the differential pressure valves that can be used for an LP shunt do not differ in their technical function from those for a VP shunt. No special technical requirements for LP valves are described in the literature and, in principle, any differential pressure valve can be used for this purpose.

However, the implantation site in the particularly dynamic lumbar region of the body results in some special requirements for the mechanical stability of an LP valve, especially for its subcutaneous fixation. The valve of a VP shunt, implanted under thin immobile scalp on solid skull bone, is subjected to much lower mechanical stress compared to the valve of an LP shunt. In the case of the LP shunt, it is essential to prevent possible migration of the valve and thus catheter breakage by fixation in the tissue. The mechanical resistance of the valve body to fracture caused by possible bending is also important here. For gravitational valves, a change in the angle of inclination (twisting) must also be ruled out, as they must be aligned exactly with the body axis. Therefore, fixation eyelets on the valve that can be sutured to the subcutaneous tissue are advantageous. Mirone et al. mention here, for example, the Integra™ Horizontal-Vertical (H-V) Lumbar Valve and the MIETHKE DUAL SWITCH Valve (DSV) as suitable valves (29). For adjustable valves, which are predominantly used nowadays, too great an implantation depth should also be avoided so as not to impair the magnetic adjustment function. In addition, a firm support or abutment should be provided in case an adjustment becomes necessary, which can be achieved, for example, by implantation on the fascia or the costal arch. Likewise, valves with an as wide as possible base area or an implantable wide-area valve mounting device can be used for this purpose. This largely counteracts the valves "sinking" into the housing during the adjustment process. Miyake used these criteria to assess the suitability of some widely used adjustable valves as LP shunts. The result is shown in Figure 7 (26).

According to Miyake's assessment in 2016, the MIETHKE proGAV® and the proSA® valve could only be used in LP shunts to a limited extent because they could not be fixed directly subcutaneously. Since 2023, however, this has been possible with the so-called " valve board", an implantable abutment made of biocompatible silicone.

In summary, based on current data, VP and LP shunts can be assumed to be equivalent in terms of performance and safety in the treatment of communicating and normal pressure hydrocephalus. The use of LP shunts has increased significantly in the last decade and extensive clinical experience is available. Lumboperitoneal shunts thus offer an alternative in the treatment of communicating hydrocephalus and related diseases - more variety that allows the treating physicians to address their patients' circumstances and concerns. In comparison, the LP shunt shows advantages as well as disadvantages and its use must be weighed individually by the responsible medical personnel. Taking into account some mechanical constraints regarding fixability and stability, in principle any differential pressure valve can be used as an LP shunt. In contrast, this does not apply so generally to all antisiphon devices, but only to gravitational valves and possibly flow reducers. In children in the growth phase, however, an LP shunt generally appears unsuitable due to the risk of spinal growth disorders.