**2. Ventilation and perfusion imaging**

In healthy patients, regional ventilation and perfusion match each other to optimize gas exchange (Fig 1). In diseased patients, changes in distribution of ventilation or perfusion or both are common. Vascular occlusive diseases, like PE, cause perfusion defects in conformity with pulmonary circulation while ventilation stays intact in these areas. This disconcordant ventilation/perfusion pattern, so called mismatch, provides the basis for PE diagnosis. Ventilation might be disturbed in acute PE due to bronchial constriction (Giuntini, 2001) but perfusion defects are usually observable in other areas (Fig 2a).

Ventilation commonly shows disturbances in lung diseases like pneumonia, tumours and obstructive diseases. Such perfusion patterns are essential to recognize as they provide additional specificity and significance to observed conditions.

#### **2.1 Ventilation**

For the ventilation study, gases may be used as they are distributed strictly according to regional ventilation. The gas that is used for V/P SPECT is metastable 81-krypton (81mKr). Its short half life, 13 s, implies that it disappears from the alveoli by decay at a much faster rate than by exhalation. Therefore, after a few minutes of breathing the test gas, the alveolar

Efficient and effective diagnostics for PE and other diseases should meet the following basic

Fig. 1. V/P SPECT images of a patient with normal ventilation (V) and perfusion (P).

(Giuntini, 2001) but perfusion defects are usually observable in other areas (Fig 2a).

In healthy patients, regional ventilation and perfusion match each other to optimize gas exchange (Fig 1). In diseased patients, changes in distribution of ventilation or perfusion or both are common. Vascular occlusive diseases, like PE, cause perfusion defects in conformity with pulmonary circulation while ventilation stays intact in these areas. This disconcordant ventilation/perfusion pattern, so called mismatch, provides the basis for PE diagnosis. Ventilation might be disturbed in acute PE due to bronchial constriction

Ventilation commonly shows disturbances in lung diseases like pneumonia, tumours and obstructive diseases. Such perfusion patterns are essential to recognize as they provide

For the ventilation study, gases may be used as they are distributed strictly according to regional ventilation. The gas that is used for V/P SPECT is metastable 81-krypton (81mKr). Its short half life, 13 s, implies that it disappears from the alveoli by decay at a much faster rate than by exhalation. Therefore, after a few minutes of breathing the test gas, the alveolar

requirements:

• Feasibility for all patients

• Low radiation dose

• Fast procedure and prompt availability of results

Techn = Technegas, MAA = Macroaggregated albumin

additional specificity and significance to observed conditions.

**2. Ventilation and perfusion imaging** 

**2.1 Ventilation** 

• Utility for selection of treatment strategy • Suitability for follow up and research

• High diagnostic accuracy and few non-diagnostic reports

Fig. 2. A patient with acute PE. Sagittal ventilation (V), perfusion (P) and V/P quotient (V/Pq) images of the right lung. **A)** At the initial examination, segmental perfusion defects are seen (arrow). The V/P mismatch is clearly delineated on V/P quotient images which improves visualization. Reduced ventilation is observed in posterior parts of the lung, where perfusion is preserved (blue arrows). **B)** Normalization of ventilation (blue arrows and perfusion is seen already after three days.

concentration will reflect alveolar ventilation. Ventilation is performed during continuous breathing of this gas. 81mKr has higher gamma energy than 99m-Technetium (99mTc) (191 compared to 140 keV) allowing simultaneous imaging of ventilation and perfusion. 81mKr is diluted from a rubidium generator that has a half life of 4.6 h. Its availability is limited and it is too expensive for general use.

Routinely in clinical practice, inhalation of a radio-aerosol is used for ventilation scintigraphy. Aerosol particles are liquid or solid. The size of the particles is of critical importance. Particles larger than 2 μm are deposited in large airways. Smaller particles are deposited by sedimentation and diffusion in small airways and alveoli. Particles smaller than 1 μm, are mainly deposited in alveoli by diffusion. Aerosol deposition is modified by flow pattern. High flow rates at forced breathing patterns and turbulent flow enhances particle deposition in airways and increases the likelihood of hot spots formation on ventilation images, particularly in Chronic Obstructive Pulmonary Disease (COPD).

Diethylenetriaminepentaacetic acid labeled with technetium, 99mTc-DTPA, is the most common agent used for ventilation scintigraphy. It is soluble in water and the size of the molecule is 492 Dalton. The average size of particles after nebulization is at best 1.3 to 1.8 μm. Due to the water solubility, particle size tends to increase during inhalation and to agglutinate in cases of bronchial obstruction where there are turbulent flows; this leads to the creation of hot spots. Because of the water solubility, 99mTc-DTPA particles also diffuse through the alveolo-capillary membrane to the blood. In a healthy patient, clearance of 99mTc-DTPA occurs with a half life of about 70 minutes. Increased clearance, leading to a shorter half life is observed where there is alveolar inflammation for any reason, such as alveolitis of an allergic or toxic nature and even in smokers. Clearance of 99mTc-DTPA can for diagnostic purposes be measured at a routinely performed V/P SPECT.

Quantitative Ventilation/Perfusion Tomography:

especially when 99mTc-DTPA is used.

patterns indicative of PE and other diseases.

al., 2009a, b), PE is reported if there is:

pulmonary vascular anatomy.

No PE is reported if there is:

of mismatch

**2.5 Presentation of V/P SPECT** 

other pulmonary diseases.

**3. Interpretation** 

**2.4 Reconstruction and calculation of V/P quotient images** 

option, particularly for quantification and follow-up of PE patients.

**3.1 Criteria for acute PE according the european guidelines** 

media or co-morbidity.

The Foremost Technique for Pulmonary Embolism Diagnosis 189

During the examination, the supine patient carefully maintains the position between ventilation and perfusion acquisitions. Immobilization for only 20 minutes is well tolerated by nearly all patients. Examination in supine position is comfortable even for critically ill patients. It is also more convenient for the staff. It is noteworthy that V/P SPECT can be performed in all patients, since there is no contraindication related to age, radiation, contrast

Iterative OSEM is essential for SPECT reconstruction. Ventilation activity is subtracted from perfusion images. A valuable parameter in clinical SPECT is the Ventilation/Perfusion quotient, V/Pquotient (Bajc et al., 2004; Palmer et al., 2001). V/Pquotient is calculated after normalization of ventilation counts to perfusion counts. Hot spot removal is important,

V/P SPECT images are usually presented in frontal, sagittal and transversal projections, available in any modern system. The slices must be accurately aligned so that ventilation and perfusion slices match each other for correct comparison. Therefore, it is crucial to achieve this acquisition in one session with maintained body position. This is also a prerequisite for the calculation of V/Pquotient images, which greatly facilitates identification of ventilation/perfusion mismatches typical of PE as well as other patterns characteristic of

Volume rendered images, such as "Maximum Intensity Projection" are available with almost all SPECT systems, allowing rotating 3D views. This function is another valuable

According to the new European guidelines, the holistic interpretation of lung SPECT is recommended (Bajc et al., 2009b). The clinician can only benefit from reports, which clearly express the presence or absence of PE. This goal was not achieved with previous probabilistic reporting methods according to PIOPED or modified PIOPED. Large V/P SPECT studies show that this is achievable if all patterns are considered, where these combine ventilation and perfusion. Conclusive reports were given in 97 to 99 % of studies. Holistic interpretation of V/P SPECT should be based upon: a) Clinical pre-test probability and b) the application of criteria for interpreting V/P patterns to distinguish between

In accordance with the guidelines of the European Association of Nuclear Medicine (Bajc et

• V/P mismatch of at least one segment or two sub-segments that conforms to the

• matched or reversed mismatch V/P defects of any size, shape or number in the absence

• normal perfusion pattern conforming to the anatomic boundaries of the lungs,

Technegas is a newer solid aerosol with extremely small carbon particles, 0.005-0.2 μm, labeled with 99mTc which are generated in a high temperature furnace. The small particle size implies that they are distributed in the lungs almost like a gas and are deposited in alveoli by diffusion (James et al., 1992). Technegas provides images which are equivalent to those with 81mKr. Technegas significantly reduces problems of central airway deposition and peripheral hotspots. Patients routinely admitted for V/P SPECT and a group of patients with known COPD were recently studied with both 99mTc-DTPA and Technegas showing superiority of the latter (Jögi et al., 2010). Unevenness of radiotracer deposition and degree of central deposition were significantly reduced with Technegas, particularly in the obstructive patients (Fig 3). In some patients, mismatched perfusion defects were only identified using Technegas because the significant peripheral unevenness of 99mTc-DTPA obscured mismatch. PE might have been overlooked in COPD patients using 99mTc-DTPA. In a few patients, 99mTc-DTPA yielded images of very poor quality. Technegas is therefore recommended as the superior radio-aerosol, particularly in patients with obstructive lung disease. A further advantage of Technegas is that relatively few breaths are sufficient to achieve an adequate amount of activity in the lungs.

Fig. 3. Comparison between 99mTc-DTPA and Technegas ventilation studies in a patient with COPD.

#### **2.2 Perfusion**

Perfusion scintigraphy involves an intravenous injection of radio-labeled macroaggregates of albumin (MAA), sized 15-100 μm, which cause microembolization of pulmonary capillaries and pre-capillary arterioles in amounts reflecting regional perfusion. At least 60 000 particles are required to obtain a representative activity distribution (Heck & Duley, 1974). Routinely, about 400 000 particles are injected. As there are over 280 billion pulmonary capillaries and 300 million pre-capillary arterioles, only a very small fraction of the pulmonary bed will be obstructed. A preparation of 100 000- 200 000 particles is recommended for patients with known pulmonary hypertension or after a single lung transplant. Degradation of MAA results in its elimination from the lung within a few hours.

#### **2.3 Acquisition**

To perform V/P SPECT takes only one hour from referral to report (Bajc et al., 2004; Palmer et al., 2001). The ventilation study starts with inhalation of 25-30 megabecquerel (MBq) Technegas, usually 2-3 breaths. Immediately after ventilation SPECT, a dose of 100-120 MBq 99mTc-MAA is given intravenously for perfusion imaging.

During the examination, the supine patient carefully maintains the position between ventilation and perfusion acquisitions. Immobilization for only 20 minutes is well tolerated by nearly all patients. Examination in supine position is comfortable even for critically ill patients. It is also more convenient for the staff. It is noteworthy that V/P SPECT can be performed in all patients, since there is no contraindication related to age, radiation, contrast media or co-morbidity.
